The Origin of Roman Praenomina

STOP

Early Journal Content on JSTOR, Free to Anyone in the World

This article is one of nearly 500,000 scholarly works digitized and made freely available to everyone in
the world by JSTOR.

Known as the Early Journal Content, this set of works include research articles, news, letters, and other
writings published in more than 200 of the oldest leading academic journals. The works date from the
mid-seventeenth to the early twentieth centuries.

We encourage people to read and share the Early Journal Content openly and to tell others that this
resource exists. People may post this content online or redistribute in any way for non-commercial
purposes.

Read more about Early Journal Content at http://about.jstor.org/participate-jstor/individuals/early-
journal-content.

JSTOR is a digital library of academic journals, books, and primary source objects. JSTOR helps people
discover, use, and build upon a wide range of content through a powerful research and teaching
platform, and preserves this content for future generations. JSTOR is part of ITHAKA, a not-for-profit
organization that also includes Ithaka S+R and Portico. For more information about JSTOR, please
contact support@jstor.org.

THE ORIGIN OF ROMAN PRAENOMINA.
By George Davis Chase.

I. The Indo-European Name.

THE brilliant work of Dr. August Fick, in his investigation of
Greek personal names,1 has furnished us with very clear and
definite ideas of an Indo-European name system and of the substan-
tially intact preservation of the same system in the great subdivisions
of the Indo-European family. Only in the cases of the Italic and
the Balto-Slavic branches does he fail to recognize the original
system, — in the former, because it has been overgrown and obscured
by an entirely local development, in the latter, from lack of evidence.
An important task remains, — and this will be the object of our
study, — to inquire whether there exist on Latin ground any vestiges
of the original system, and to consider in what way the historical
Latin order of names may have been developed.

It is necessary, first and foremost, to understand exactly the
principles which Fick has established. They are, in so far as they
bear on our problem, in the main these : Each individual, man or
woman, had a single name composed of two members. Familiar
examples of these are Gk. A^o-o-^enjs, A vest. Atared&ta (='A.Tpa-&aT-qs,
from A tar, the sacred fire), Skt. Brhad-afva (=' Great-horse '),
Germ. Fridu-mar, Celt. Dumno-rix (from dumno — dubno, Goth, diups,
'deep'), Slav. Vladi-mir. That exactly the same kind of name was
used in Balto-Slavic as well, has since been proved for the Lithuanian
and Old Prussian (Brugmann, Vergl. Gram, ii, § 18), and is shown
for the Lettish by a number of names of heroes from heathen times
which I have been able to discover. Thus Tali-walds ' Wide-ruler,'
from tali ' far ' (= Gk. rrj\e, cf. TyXi-frnxos, etc.), and waldit, Germ.

lDie Griechischen Personennamen, etc., 1874; second enlarged edition, by Fick
and Bechtel, 1894.

104 George Davis Chase.

walten; Wisu-walds 'All-ruler,' from wiss 'all,' Skt. vicva, Slav, vhi,
of which the exact counterpart is found in the Slavic Vse-volod, from
vM and vladiti; Ldf-plesis ' Bear-tearer,' from lads 'bear,' and plest
' to rend.' Perhaps also the mythical hero Margers (cf. Latin Mars),
who fell in love with Maija, the goddess of love, and whose exploits
are celebrated in the poetry of the Lettish people, also exhibits an
example of a similarly formed name.

Most commonly each member was of two syllables, but often of
more, as 'AiroAA.o-8a>pos, and not infrequently of a single syllable, as
Ev-KpaTrp. That both members of a name, however, should, in the
original form, be single syllables, is a rare occurrence, although in
Celtic, Germanic, and Slavic, laws of unaccented syllables have often
reduced names to dissyllabic form. Thus such names as IleA.-oi/' are
uncommon, but names like Germ. Dag-frid, for *Daga-fridu, are
frequent.

It is evident that in any language an equipment of names averag-
ing four syllables in length, while possessing dignity and sonancy,
and appropriate to the requirements of epic narrative, heroic
tradition, and formal intercourse, is ill-adapted to the needs of
familiar, every-day life. A shortening and simplification is the inevi-
table result, and this is amply illustrated in each of the Indo-European
tongues. The process was in nowise different from that which we
see going on about us every day : Frederick is familiarly called Fred,
Catherine, Kate, until these shortened names become so familiar that
they are received into the class of given names, and we meet with
men and women who were never named anything but Fred or Kate.
Further, Elizabeth may be shortened to Eliza, Lizzie, Betsey, Betty,
or Bessy, and so, from one original, our list of available given names
is considerably increased. It is to be noticed, furthermore, that
either part of the original name may be taken as the shortened form ;
thus Joseph is in English abbreviated to Joe, in Swiss to Seppi; Fri-
derike becomes either Frida or Hike. Again, Lou may stand for
Louise, or Lucy, or Lucinda ; Bert for Al-bert or Bert-hold. In like
manner the early compound names were abbreviated ; either member
or the first member and the initial syllable of the second might be
taken as the new name. Since the same member might form a com-
ponent part of a number of names, it might stand as an abbreviated

The Origin of Roman Praenomina. 105

form for any one of them. Thus Fick cites sixty-one names begin-
ning with Wulf, and an equal number ending with it. The shortened
form Wolfo, or Woljin, or Wolfilo may be the representative of any
or of all of these hundred and twenty-two names.

To the shortened names were added suffixes of a considerable
variety of form, but in force mainly diminutive, usually with the addi-
tional idea of endearment. Among the commonest suffixes are -*-, as
in Gk. Kvirp-i-s, Hdp-is, Germ. Ruodi, from Ruod-olf, etc. (cf. Eng.
Billy, Johnny, etc.) ; -ia-, as in Skt. Dev-iya, from Deva-datta, etc.,
and very commonly in Greek, as in Nuc-i'a-s : -ino-, as in German
Wolf in, etc. : and also suffixes having /, k, or d as their character-
istic, as Skt. Devi-ka, Gk. "lwira-Kos, Avk-iS-tjs ; Germ. Ulf-ila, Wolf-izo,
Frizzo (=*Frid-td).

To illustrate further the formation of the compound names and the
simplified ones derived from them, we will give a few of the more
common examples :

Greek, — Xpvcr-ayopas, Xpv<ro-yovos, Xpv<T-wnros, Xpucrd-jaaxos, Xpvcro-
yevrj's, Xpvo-d-0e/us, etc. ; XpScros, Xpwnov, Xpwis, Xpvcrias, XpwiW, etc. ;
TrjX-avyrji, TijXtryovos, TtjXc-SCki), TjjAe-KAuTOs, TijA.e-jU.axos, Tr]X€-<j>dvj}i,
TrjXe-viKos, TiyA.€-p./3p0T0S, etc. ; TjjAeas, Tr/Xrjs, TrjXivrjS, TrjXiav&rjS, etc.

Sanskrit, — Mitra-deva, Mitra-bhanu, Mitra-varcas, Mitra-vinda,
Mitra-sena, Mitra-(arman, etc.; Mitra, Mitraka, etc.; Rupa-dhara,
Rupa-manjari, Rupa-fikhd, Riipa-sena, Kama-rupa, Su-rupa, etc.;
Rupa, Rupin, Rupya, etc.

Germanic, — Fridu-berht, Frid-wald, Frid-dag, Fridu-ger, etc. ;
Muot-frid, Mar-frid, Arn-frid, Liub-fried, etc. ; Frido, Fridiko, Fricco,
Fridilo, Frillo, Fridolin, Fridin, Frizzo.

Celtic, — Touto-bocio, Toutio-rix, etc., from Ir. tuath (= Goth, \iudd) ;
Toutus, Touta, Toutia, Toutillus, Toutona, Toutonia.

Avestan, — Kereftifpa 'lean horse', Manus-cithra (cithra 'bright'),
Nairyo-fahha (nere 'man').

Slavic, — Ljude-vit, Ljude-mysl (Lj'udu ' people ' ) ; Ljuda, Ljuden,
Ludek ; Dobro-voj, Dobro-gost (dobru 'good ' ) ; Dobr, Dobrilo, Dob-
ren, Dobrota, Dobrik, Dobros.

106 George Davis Chase.

There is abundant evidence to prove that the shorter forms are
abbreviations from compound names, and not separate names. On
this subject, compare Brugmann (Vergl. Gram, ii, t>Z) for a clear
and concise statement of the facts. For the Germanic, Fr. Stark
(Die Kosenamen der Germanen, Vienna, 1868 ; cf. Fick1, xcii) has
collected from documentary evidence a great number of examples in
which the same person is called both by the full name and the
shortened form. Thus, in Icelandic, we find Kostbera in Elder
Edda, Atlam&l, 6 and 9 ; but ibid. 33 and 50 the same person is
mentioned as Bera or Beru. In Sanskrit, the grammarian Panini
(5, 3, 78-9) recognizes the abridgment of names as a regular
principle, and gives rules of grammar for the formation of the
shortened forms. Examples, too, are not uncommon in Sanskrit
literature. Professor C. R. Lanman has called my attention to
a number of instances, as follows: MBhr. iii, 15582 Kotika-asya
(Prince) 'Frog-mouth'; ibid. 15586, shortened to Kotika 'Frog.'
(Cf. Lettish Kaupis ' Frog', the name of a hero of heathen times.)
Jataka, ed. Fausb^ll, i, 241, 3, Mitta-vindaka ; ibid. 241, 9, Mittaka.
Sutta-nipata (p. 61, line 5), Nigrodha-kappa, called for short Kappa,
ibid. (p. 62, stanza 349). In Greek, the grammarian Herodianus
(Etym. M. 93.50) has the following passage : "A^is • tovto ov
(TvyKoirij, aXXa. /i£Tacr^i;/*oTiO"jtios. airo yap tov ' Ap-fadpaos "A/x<f>t's <us
Trap Al<r\v\w, tixnrzp I<piavacrcra I<f>fs /cat ®pa<rvK\ijs ©pdVvAAos Kal
Ba6vK\r)s Ba0vX\os vTroKopioriKa. Plato (Protag. 318, b) speaks of
Zeuxippus of Heraclea as a famous painter ; this is generally under-
stood to be the artist who is often mentioned by the name of Zeuxis
(cf. Plat. Gorg. 453 c ; Xen. Mem. i, 4, 3, etc.).

One more essential in connection with the Indo-European name
system demands our attention, — the character of the words from
which the members are derived. It was no part of the design of the
progenitors of our race to attach to their descendants names that
would involve a reproach or encumbrance. Rather, they sought, as
omens for virtue and prosperity, to endow them with appellations
suggesting divine favor and good will, or descriptive of their con-
ceptions of ideal men and women (see Laws of Manu, ii, 3 1 seq.).
It follows, therefore, that their compounds were made from a choice
of words as limited as their ideal conceptions of beauty, amiability,

The Origin of Roman Praenomina. 107

prosperity, and virtue. These ideas were oftenest expressed by such
nouns as war, spear, shield, peace, law, people, friend; joy, mercy,
desire; horse, deer, wolf, dragon, serpent ; by such adjectives as good,
long, high, far, all, strong, true, bright, renowned; by such verbs as
know, support, defend, beget, shine, possess, do, stand, destroy, fight,
rule. The character^and limitation of these words is easily recog-
nized, and is a most important feature of the system.

II. Latin Names.

Having once grasped the full significance and the extension of
this common Indo-European name system, we cannot for a moment
doubt that it once existed on Italic soil ; but when we look for its
traces we are confronted with obscurities. It may well be ques-
tioned whether a single survival of the compound names exists in
Latin. Publi-cola is the nearest approach to a probable instance,
and the probability is heightened when we compare similar names
in other languages, such as Gk. Aij/to-i^tXos, *t\d-8^os ; Slav. Ljude-
vit ; Irish Tuath-char (Touto-carus) ; A.S. Lead-win. But we are
obliged to regard this correspondence as illusory, for we have the
distinct tradition (Liv. ii, 8, 1) that Publicola was added as a cog-
nomen to P. Valerius in 509 B.C., because of his efforts in promot-
ing the republic ; that is, he was the man who ' worked the people.'
Plebi-cola is apparently a later name formed on the analogy of Publi-
cola. Aheno-barbus and Crassi-pes are old compounds, but from their
signification seem to belong to the Italic rather than the Indo-Euro-
pean system. Centum-alus has been conjectured to mean ' one who
supports an hundred slaves ' (from centum and alere), but its deriva-
tion is entirely unknown and many other possibilities may be
thought of. Opiter, though often believed to be a compound, is so
obscure that we must reserve the discussion of it till later.

It is plain, therefore, that if we are to find in Latin remnants of
the original name system we must seek for them in the abbreviated
forms. For this reason I have been particular to call attention to
the characteristic features of the shortening process and to the
nature of the members of which the name is composed. These we
may hope to find valuable guides in our investigation.

108 George Davis Chase.

It is possible that the Italic peoples, at some prehistoric time,
from some powerful outward pressure, may have utterly abandoned
their own name system for a foreign one. Such an occurrence is
not unknown in the history of races, as, for instance, when a
heathen people embraces Christianity and is baptized with Christian
names. Thus the Lettish people of Courland, when, several cen-
turies ago they were at once converted to Christianity and reduced
to slavery by the Prussian nobility, laid aside forever their ancestral
heathen names and became at the hands of the priesthood simple
Peter and John. Such a change may be so sudden and complete as
to leave no traces of the earlier practice. That such, however, was
the case among the Italic peoples is not at all probable. Far more
likely is it that out of the old system, and building upon its ruins,
there was developed the new system which we meet in historical times.
Nor is the reason for the decay of the old far to seek. It rested upon
the fundamental abhorrence of the Latin language for long compounds.
We have seen that the Indo-European names average four syllables
in length. Such compounds we easily recognize as foreign to the
tendencies of Latin. The result was that the compounds were early
abbreviated and continued to exist only in the shortened forms.

But we are basing this argument too largely on the inferences of
logic without the reinforcing evidence of the facts. It is now time
that we examine carefully the Roman system in its component parts.
In republican times, at least, the Romans themselves recognized and
clearly distinguished in a man's complete official name three divi-
sions, which were written in a regular, definite order. First, the
praenomen, or name given to the child by its parents on the ninth
day after birth (cf. Marquardt, Rom. Alterth. vii, 10); secondly,
the gentilicium, or name from the gens into which the child was
born ; thirdly, the cognomen, or name of the particular family to
which the child's father belonged, or the name which the child
might acquire in after life from achievement or peculiarity, and
which would be handed down to his children. Thus a man might
be born to one or more cognomina and gain others during his life-
time. This happened particularly in aristocratic families, where we
find, even in early times, such names as P. Cornelius Scipio Nasica
Serapio, the consul of 133 B.C.

The Origin of Roman Praenomina. 109

III. Latin Cognomina.

The cognomina, as in origin and structure the most transparent
class of Latin names, may first claim our attention. To consider
every cognomen in Latin would be a task beyond the scope of the
present work, and indeed is not necessary to any understanding of
the cognomina as a class. For our purpose it has seemed sufficient
to examine and classify all the cognomina from a considerable portion
of the Latin sources. As of sufficiently wide range and sufficiently
representative of the whole, we have chosen all those occurring in
Livy and in the first volume of the Corpus Inscriptionum Latinarum
(CIL), exclusive of those names, confined to slaves, which are mani-
festly of foreign origin. These we have classified into the groups
into which they naturally fall according to their meaning.

We have first a list of names which refer to physical peculiarities,
as follows: Capito 'big-head,' from caput; Cotta 'cowlick' (?) 1 ;
Nasica 'prominent nose,' from nasus ; Naso 'large nose'; Silus
'pug-nosed' ; Silo 'pug-nosed,' from silus ; Licinus 'bent upward'
(of the nose ?) ; Mento ' long chin,' from mentum ; Dentio, Denter,
'prominent teeth,' from dens; Brocchus 'prominent teeth' ; Labeo
' thick lips,' from labrum ; Chilo ' large-lipped,' from x«ta>* (cf. Fest.
43. 10); Flaccus 'flop-eared' ; Strabo 'squinter' ; Codes 'blind in
one eye,' for Kikf ; Paetus, Paetinus ' blink-eyed ' ; Luscus, Lusci-
nus ' one-eyed ' ; Pronto, Cilo ' prominent forehead ' ; Bucca ' big
cheeks,' or ' bawler ' ; Scaeva (sc. manus) ' left-handed ' ; Scaevola,
diminutive of the same; Laevinus 'left-handed,' from laeva ; Laeca
' left-handed,' for Laevica (sc. manus') ; Unimanus ' one-handed ' ;
Varus ' knock-kneed ' ; Valgus ' bow-legged ' ; Sura, Sulla (for

1 The explanation of Cotta is very difficult. We may understand it to be con-
nected with the Sicilian k6ttos, kottIs meaning ' forelock, cockscomb ' (?) (Hesy-
chius, irp6K0TTa- elSos Kovpas, rj K«pd\rjs, Tplxupui- kottIs yi.p ij /ce0a\i). ical ol d\e-
Krpv&ves kottoI Silt. rbv M t5 (ce^aXj \&<pov. Cf. Sartori, Kottabos-spiel, 1893, p. 73).
Cotta is otherwise explained as standing for Coda ; cf. Wolflin, Archiv f. Lat. Lex.
vi, 269. The assimilation of -ct- to -tt- is not Latin, but is regular for modern
Italian and may have been dialectic in classical times. It might then refer to a
sun-burned face, ' parboiled.'

HO George Davis Chase.

surula, a diminutive of sura l ) ' prominent calves ' ; Scaurus
' swollen ankles ' ; Crassipes ' thick-foot ' ; Plautus ' flat-foot ' ; Plancus
' flat-foot ' ; Pedo ' broad-foot,' from pes ; Pansa ' broad-foot ' ;
Falto (= Baled) ' a person with the great toes turned in ' ; Dossenus,
Dorso ' hunchback,' from dorsum ; Coxsa ' prominent hips ' ; Galba
' fat-paunch,' a Gallic word (cf. Suet. Galba, 3) ; Mammula ' little-
breast,' from mamma ; Scapula ' prominent shoulders ' ; Balbus,
Blaesus ' stammerer ' ; Blasio ' stammerer,' from blaesus (?) ; Rocus
'hoarse,' for raucus ; Tubero, Tubertus 'bumpy,' from tuber; Verru-
cosus 'warty' ; Macatus 'spotted,' a past participle of a *macare, a
denominative verb from * maca, of which the diminutive macula is
preserved; Mancinus ' maimed,' from mancus ; Peticus 'scabby' (?)
(cf. petigd) ; Callus 'thick-skinned,' or perhaps for Gallus ; Nerva
'sinewy,' from nervus ; Barbatus 'bearded'; Barba 'long-beard';
Barbo ' long-beard,' from barba ; Ahenobarbus ' bronze (colored)
beard' (cf. Suet. Nero, 1) ; Comatus ' long-haired ' ; Caesar 'hairy'
(cf. caesaries, Skt. kef a) ; Crispinus ' curly hair,' from crispus ; Cincinna-
tus ' curly hair ' ; Cinnus ' curly hair,' (cf . Cinna, sc. coma) ; Elva or
Helva (sc. coma) ' light bay hair ' ; Lanatus ' woolly ' ; Glabrio ' bald-
head,' from glaber ; Calvus 'bald' ; Calvinus 'baldish,' from calvus ;
Vulso ' plucked,' from vello ; Rufus ' red-haired ' ; Rufinus ' reddish-
haired,' from rufus ; Rufio 'red-haired,' from rufus ; Rutilus 'red-
haired' ; Flavus ' fair-haired, Fairfax'; Albus 'white (haired?)';
Albinus ' whitish (haired ?) ' ; Purpureo ' rosy (complexioned) ' ;
Niger ' dark ' ; Fldrus ' shining (complexioned) ' ; Ambustus ' sun-
burned ' ; Curio ' emaciated ' ; Pennus ' sharp (featured) ' ; Macer
'lean' ; Macerinus ' leanish ' ; Crassus ' fat ' ; Bassus ' stout ' ; Paullus
' short ' ; Longus ' tall ' ; Longinus ' longish ' ; Pulcher ' handsome ' ;
Curvus ' bent ' ; Drusus ' stiff.' 2

A second class refers to the habits or character of the individual,
as :

Casca ' old-fashioned,' from cascus ; Rusticus ' living in the country,
countrified'; Alienus ' foreign ' ; Cato ' sagacious,' from catus ; Nero
' noble ' (a Sabine word connected with nerio, fortitude, Skt. nara,

1 For a different explanation of Sulla, see Plut. Sulla, 2.

2 For another explanation of Drusus, see Suet. Tib. 3.

The Origin of Roman Praenomina. 1 1 1

Gk. Avijp ; in Oscan nero seems to be a title of rank) ; Severus
' stern '; Violens ' furious ' ; Lepidus ' agreeable ' ; Frugi ' temperate '
Magnus ' great '; Pius ' dutiful ' ; Carus ' dear ' ; Imperiosus ' haughty '
Augustus 'majestic'; Nobilior 'nobler'; Molliculus 'voluptuous'
Venox ' given to hunting ' ; Lentulus ' rather slow ' ; Celer ' swift '
Fessus ' wearied ' ; Brutus ' coarse ' ; Tremulus ' shaking ' ; Varro
and Varo (= Baro 1) ' a foolish person ' ; Regillus ' royal ' ; Potitus
' boss ' ; Structus ' built up ' ; Bibulus ' tippler ' ; Bibaculus ' tippler,'
a diminutive of bibax ; Tappo, Tappulus ' tippler ' (cf. lex tappuld) ;
Dives ' rich ' ; Gurges ' spendthrift ' ; Rullus ' beggar ' ; Sabula ' a
talkative person' ; Publicola ' one who works the people' ; Plebicola
' one who works the plebs ' ; Caldus ' warm ' ; Acidinus ' sharpish.'

A third class refers to the condition or relation of the individual,
as :

Faustus ' fortunate ' ; Fostlus = Faustulus, a diminutive of Fau-
stus ; Felix ' fortunate ' ; Donatus ' gifted ' ; Auctus ' increased ' ;
Liber ' free ' ; Spurinus ' illegitimate,' from spurius ; Cordus ' born
late in the season'; Proculus, a diminutive of Proeus;1 Primus
'first'; Septimus 'seventh'; Postumus 'last born'; Maximus
' eldest ' ; Priscus ' old ' ; Geminus ' twin ' ; Gemellus ' twin ' ; Trige-
minus 'triplet' ; Frater 'brother' ; Paterculus 'daddy.'

The fourth class is of offices and occupations, or names derived
from such, as :

Figulus ' potter ' ; Pictor ' painter ' ; Fullo ' fuller ' ; Natta or Naeta
' fuller,' from voo-o-oj, to press ; Faber ' smith ' ; Lanio ' butcher ' ; Llbo
'sprinkler,' connected with libare;2 Subulo 'flute player' ;,Thurarius
'frankincense dealer'; Bubulcus 'cowherd' ; Pictor 'baker of offer-
ing cakes'; Cursor 'runner'; Mensor 'measurer' ; Pollio 'polisher
of arms ' ; Cornicen ' trumpeter ' ; Nauta ' sailor ' ; Metellus ' mercen-
ary,' = fufr6tos (cf. Paul, ex Fest. p. 146) ; Triarius, a soldier serving
among the triarii; Salinator ' salt-dealer' ; Camillus, a noble youth

1 For another explanation of Proculus cf. Paul, ex Fest. 225, who explains it as
meaning ' born when one's father is abroad.'

2 The relation between Llbo and libare may be explained as one of ablaut,
libarc standing for leibarc ; cf. feido and /Ides. The name Libo may come from
the office of libation-pourer at the sacrifice.

112 George Davis Chase.

employed in religious offices ; Flamen; Flamininus, from flamen ;
Augurinus, from augur; Censorinus, from censor; Rex 'ruler';
Regulus 'prince ' ; Hortator 'exhorter' ; Flaccinator, from flacceo;to
flag, 'one who makes to flag'; Sulca .' furrower,' from sulcare, to
furrow ; Numilor ' arranger ' (cf. Numa, vo/u.os, numerus, etc.).

The fifth class, a large and very characteristic one, consists of the
names of common objects, animals, etc. Names also derived from
animals and objects, as Laenas from laena, Torquatus from torques,
Caepio from caepa, may best be classed here, as the difference
between calling a man ' necklace ' and ' wearing a necklace ' is purely
one of rhetorical figure. The relationship of all these names to the
individual is often not clear ; but the general character of the rela-
tionship is sufficiently suggested by those examples concerning which
there is no doubt. In the case of objects directly connected with the
person, as Crista, Spinther, Torquatus, the relation is easy to see ;
in the case of animals, birds, etc., as Asinus, Vitulus, Merula, the
name doubtless ascribed to the person the character or habits for
which the animal is supposed to stand. So in English, and particu-
larly in the zoological ' Schimpf worter ' in German, the names of
animals are regularly applied to persons, usually in an uncomplimen-
tary sense. In Latin the kind of person designated by many of
these animal names is no longer known, but from those which we do
know the general intent of . the class is not hard to infer. The most
difficult subdivision of all is that of inanimate objects, in some of
which a resemblance to the individual, in appearance or character,
seems to be hinted at, as in Tubulus, Culleo, Stola; in others the
person may have had some real business or connection with the
object, as in Carbo, Buxsus. It is very difficult to decide, for ex-
ample, whether Caepio, from caepa ' onion,' means a man who culti-
vated onions, or whose head was shaped like an onion. Perhaps for
our purposes of classification it is sufficient to know that he was nick-
named ' oniony.' The list is as follows :

Spinther ' bracelet ' ; Crista ' crest ' ; Centho, a cap worn over the
helmet; Per a ' wallet ' ; Fimbria 'fringe'; Laenas 'cloaked,' from
laena (cf. Cic. Brut. 14) ; Argentillus, from argentum, silver; Tor-
quatus ' adorned with a torques ' ; Asellus ' little ass ' ; Asina ' she
ass ' ; Catulus ' puppy ' ; Caninus ' doggish ' ; Vitulus ' calf ' •, Trio

The Origin of Roman Praenomina. 113

'plow ox'; Vaccus, from vacca, cow (?) ; Verris 'male swine' (cf.
Plaut. Mil. 1059, where it is used contemptuously of a man) ; Mus
' mouse ' ; Murena, a shellfish (referring to the color of his dress) ;
Stellio ' newt, crafty fellow ' ; Merula ' blackbird ' ; Corvus ' raven ' ;
Turdus ' thrush ' ; Buteo ' hawk ' ; Gragulus (= Graculus) ' jack-
daw ' ; Gracchus 'jackdaw,' for * gracus, of which graculus is the
diminutive ; Todillus, a kind of small bird (cf. Fest. p. 352) ; Musca
' fly ' ; Cossus, larva under the bark of trees ; Caepio, from caepa, an
onion ; Cicero, from cicer, a small pea ;l Fabatus, iromfaba, a bean ;
Fundulus ' sausage ' ; Aculeo, from aculeus, a spur ; Malleolus ' small
hammer ' (cf. Charles Martel) ; Dolabella ' little pickax ' ; Tegula
' tile ' ; Tubulus ' small tube ' ; Piso ' mortar ' ; Gillo ' cooling vessel ' ;
Culleo, from culleus, a leathern bag ; Maso or Masso, from massa, a
lump (cf . Massa, a cognomen) ; Scipio ' staff ' ; Buxsus ' boxwood ' ;
Carbo ' coal,' also used of a thing of small value ; Pulvillus ' little
pillow'; Stolo 'branch, sucker'; Ahala i=Ald) 'wing'; Corbo,
from corbis, basket ; Saxa ' rocks ' ; Saxula ' little rocks ' ; Arvina
' grease ' ; Lucullus, from lucus, a grove (a double diminutive) ;
Ralla (= raduld) ' scraper' (cf. rallum) ; Cerco, from Kepxos 'tail ' (?) ;
Posca, a drink of vinegar and water ; Scylla, the sea monster ;
Merenda ' afternoon luncheon ' ; Musa ' muse ' ; 2 Limetanus, from
limites; Alimentus, from alimentum, nourishment.

The sixth class includes names from localities and represents the
town, district, country, or people from which the person came or
with which he was in some way identified, as :

Antias, from Antium ; Fidenas, from Fidenae; Sufenas, from Sufena;
Asprenas, Aesillas, Supinas, from unknown towns ; Menas (written
Mend), Umbrian, and perhaps a gentile name ; Fregellanus, from
Fregellae ; Tusculanus, from Tusculum ; Camerinus, from Cameria;
Medullinus, from Medullum ; Coriolanus, from Corioli; Setinus, from
Setia ; Norbanus, from Norba ; Collatinus, from Collatia ; Regillensis,
from Regillus ; Nbmentanus, from Nomentum; Pyrgensis, from

1 An interesting parallel to Cicero is the Sanskrit personal name CSnakya,
from canaka ' cicer.'

2 The cognomen Musa is derived from mussus according to Fisch, die Latein-
ischen nomina personalia aufu o, onis," 1890, p. 157.

114 George Davis Chase.

Pyrgi; Veientanus, from Veii; Calenus, from Cales ; Nblanus, from
Nola; Atellanus, from Atella; Seranus, from Serranum; Caudinus,
from Caudium ; Tempsanus, from Temesa; Tarentinus, from Taren-
tum; Sarranus, from Sarra; Massiliota, from Massilia ; Atratinus,
from Atratus, a river near Rome; Silanus, from 6V/a, a forest in the
country of the Brutii ; Capitolinus, from the Capitolium ; Tuscivica-
nus, from Tuscivicus ; Coelimontanus, from Coelemontium ; Arsa, a
city of Spain ; Sarra, the city of Tyre ; Croto ; Istra ' Istrian ' ;
Numida, Sabinus, Lucanus, Marsus, Apulus, Umber, Volscus, Ligur,
Gallus, Siculus, Hispanus ; Hisp 'alius <* Hispanulus 'little Span-
iard'; Cerretanus, from the Cerretani ; Vetto or Vecto, from the Vec-
tones ; Geta, from the Getae; Surus 'Syrian'; Creticus, from Crete;
Isauricus, from the Isauri; Africanus, from Africa; Asiagenus
(= 'Awiaywfc) 'born in Asia': Messala, from Messana (cf. Sen-
Brev. Vit. 13, 5) ; Mugillanus, Tuditanus, Vibulanus, Turrinus, Tri-
cipitinus, Pontenus, from unknown places.

The seventh class consists of names derived from other names.
The majority are names in -anus, from gentile names, and denote, in
the case of an adopted son, the gens in which he was born. The
others are plain derivatives from familiar names, or are names of
divinities applied to men. The list is as follows :

Clodianus, Nonianus, Cestianus, Vinicianus, Octavianus, Caeicianus,
Aemilianus, Sextianus, Messianus, Sempronianus, Petrinianus, Rulli-
anus, Satrienus, derived from gentile names ; Saturninus, from
Saturnus; Mamerrinus, from Mamercus ; Marcellinus, from Marcellus ;
Marcellus, a diminutive of Marcus ; Lucillus, a diminutive of Lucius ;
Silenus, the companion of Bacchus ; Myrtilus, the son of Mercury.

The eighth class consists of names of foreign origin, many of
which are Greek or Etruscan, as :

Philus, 4>t\os ; Philippus, &i\tiriros ; Thermus, %e.ppo<i ; Sophus,
Solo's ; Philo, 3>tA<ov ; Matho, Md0tav ; Molo, MoAtov ; Gritto, Ypiaawv
' pig,' used as a personal name in Greek ; Bursio, from the cogno-
men Bursa, for fivpcra, a hide ; Tympanus, from Tvp.va.vov, a timbrel ;
Hupsaeus ioxHypsaeus, from ui/'os, height; Basilius, /Sao-tXctos ; Megellus,
in Greek M«yt\Aos (Xen. Hell, iii, 46), from p.iya.% ; Pitio occurs as
IIiTtW on a grave-tablet in the Theseion. Etruscan names are
Spurinna, Sisenna, Siperna, Perperna, Perpena, Thalna ; Oscan

The Origin of Roman Praenomina. 115

names are Papus, Paccius, Pacilus, Salvius ; Maro is Umbrian, cf.
maronatei ' praetura ' ; Matuginensis, from Matugenus, is a Celtic
name.

Tamphilus is a name of great obscurity. O. Keller, Lateinische
Sprachgeschkhte, ii, 268, attempts to derive it from Damophilus, a
process difficult both on account of the syncopation and the change
of d to /. The regular change in Greek loanwords is from initial
tenues to mediae, as gubernare < Kvficpvdw. Syncope, too, under the
old Latin accent law hardly occurs in foreign words, which were
usually borrowed after the penultimate accent law went into effect.
Tamphilus is found in the gens Baebia, a family from Oscan terri-
tory. It seems most probable that the name is connected with the
gentile Tampius, which occurs in the same part of Italy, cf. ' CIL.
ix, 5190 ; iii, 2547. Perhaps the form which appears in CIL. i, 264,
M. Baebius Tampilus, is the more correct, representing an original
Tampulus, which we may suppose was influenced by the analogy of
Greek names in -<f>i\o<;. In republican times there was little differ-
ence in pronunciation between -pulus and -<£t\os among the Romans.
We prefer, therefore, to assign Tamphilus to Oscan territory and
leave its meaning unexplained.

Occasionally we meet with such names as lulus, Tullus, and
Caius used as cognomina. These will be treated more fully in a
later chapter on praenomina, and may therefore be neglected here.
There remain to form the ninth and last class only a small number
of names which are of entirely unknown signification. But even
these are similar in appearance to those of our recognized classes.
There are none that are demonstrably different in character or origin
from those which we have already classed. They are as follows :

Centumalus, Cethegus, Rebilus, Volimus, Mateidus, Birbatrus, Sta-
bilius, Falevius, Bala, Calussa, Scato, Saverrio.

The general transparency of the Roman cognomina and the well-
defined classes which they form must be evident to all from an ex-
amination of the foregoing lists. Even if a doubt exists in the case
of a number of words as to their exact etymology, there can be little
doubt as to their general character. To make our results still more
plain we will add the following table :

Ii6 George Davis Chase.

ass I.

Physical peculiarities

95

II.

Individual habits or character

38

III.

" condition or relation

'9

IV.

Offices and occupations

3°

V.

Animals, objects, etc., by metaphor

60

VI.

Locality

64

" VII.

Derived from other names

20

" VIII.

Foreign names

27

IX.

Unknown

12

Total 365

The general tendency of those names which are distinctly Roman
and are not derived from other names or from places, is to point out
some defect in the individual. The actuating spirit of such a sys-
tem of names is one of gibe and criticism, — a spirit in every
respect opposed to that which we have shown was the basis of the
common Indo-European system. With regard to their signification
alone, hardly a single name from our lists of cognomina could be
derived from the original system. The name Nero might be, but
that we shall try to prove was not Roman, and with it "our list stops.

But there is additional evidence as to the origin of the cog-
nomina, — namely the form of the words themselves. They are
derived (1) from gentile names by means of the suffix -anus, or
from other names by the suffix -inus, as Mamercinus, Longinus, or by
the diminutive -ulus, as Marcellus for * Marcululus, Hispallus for
* Hispanulus ; (2) from places by the endings -anus, -inus, -hius,
-ensis, -as ; or (3) from objects, by the ending -dtus, as Torquatus,
Fabatus. But in the great majority of cases they are nouns or
adjectives taken bodily, and without any change, out of the spoken
Latin, as Corvus, Catulus, Pictor, Brutus, Rufus, Paetus. A consid-
erable number are derivatives in -0, -onis, originally from adjectives,
later from nouns and verbs (cf. Lindsay, Lat. Lang. p. 349, § 55),
used most frequently with a contemptuous signification. For the
development and use of these names, see Fisch, die Lateinischen
nomina personalia auf " o, onis," 1890. A few show an -r or -er
suffix, as Denter, beside Dentio. This ending seems to have a com-
parative force, denoting a ' kind ' or ' sort of.'

The Origin of Roman Praenomina. WJ

Many cognomina are in origin feminine -a stems. These are ex-
plained (Zimmermann, Wolflin's Archiv f. Lat. Lex. vi, 269) either
as feminine nouns denoting parts of the body, etc., as Coxa, Axilla,
Bucca, or as adjectives modifying some feminine noun understood.
Thus manus is to be supplied with Laeva, Curva, Pola ; coma with
Alba, Helva, Casca, Cinna, etc. Thus from the simplicity and trans-
parency of the formation of Latin cognomina, as well as from
their meaning, it is evident that they can have nothing to do with
the original Indo-European names.

IV. Latin Gentilicia.

Having established the character of the cognomina, we must next
direct our attention to the gentile names. These present many ob-
scure and perplexing details, a full discussion of which would far
exceed the possibilities of the present work. Nor is it necessary
for the purpose in hand. We shall be content if we can decide
upon the origin and character of gentile names sufficiently to draw
a definite conclusion ; and that we believe we can do without enter-
ing upon all the details of the problem, or even considering the
whole body of names which may be gathered. As a thoroughly
representative group, we have selected the names in the first volume
of the Corpus Inscriptionum Latinarum (CIE), and have undertaken
to classify and arrange them.

The gentile names of the city of Rome itself extend back to the
earliest period, and because of their antiquity are difficult to under-
stand. As the number of names indefinitely increased with the
gradual extension of citizenship throughout Italy, we find puzzling
names from non-Latin dialects or foreign languages constantly
creeping in. The result is that we have many gentilicia of whose
elucidation we must always despair. The first thing which chal-
lenges our attention in the consideration of gentile names is that
they are all derivative adjectives, formed in most cases by the end-
ing -ius, occasionally by other adjective-forming suffixes. Excep-
tions to this rule are found in a very limited number of exactly such
words as the underived cognomina, like Verres ' boar.'

Names of Etruscan origin often appear with the endings -erna,

n8 George Davis Chase.

-enna, -ena, as Perperna, Siperna, which have been previously classed
as cognomina, but which appear to be used without distinction as
cognomina or gentiles.

The Umbrians frequently employed as gentiles cognomina in -ds,
-atis, which were derived from names of towns, as Fulginas, Capidas.

From names of places we find gentiles either in -anus, in exactly
the same form as the corresponding cognomina, or with the further
derivative termination -anius. In either case cognomina in -anus
appear to lie back of them. For a full and satisfactory treatment of
these names, see E. Hiibner, Ephem. Epigraph, vol. ii, p. 25. In
CIL. i, we find the following representatives of these two classes :

Ae/o/anus, Irom the town Aefola; Lorelanus, Musanus; Trebulanus,
from the town Trebula ; Afranius ; Albanius, from cogn. Albanus,
from Alba; Atanius or Athanius ; Avianius, from cogn. Avianus,
from gens Avia ; Fundanius, from cogn. Fundanus, from the town
Fundi; Maianius, from cogn. Maianus ; Trebanius, from the town
Treba; Car anius, Turr anius, Tusanius.

A form of gentile in -enus, standing for -inus after -£, as in
alienus, — a name not Latin, but found most frequently in Picenum
and vicinity (Mommsen, De dialectis Italiae inferioris, p. 362), — seems
to be derived from other gentiles, rarely from places. Hiibner, /. c.
p. 28, has collected about two hundred examples of such names.
In CIL. i, the following examples appear :

From other gentiles, Aienus, Alfenus, Audienus, Betilienus,
Caesienus, Matienus, Mutienus ; Salvienus, from Sa/vius; Veienus,
from Veii ; Volsienus, from the cognomen Volsus.

Cognomina in -0, -onis form the basis of a considerable class of
gentiles in -onius (Fisch, die Lat. nom. pers. auf -0, -onis, p. 156).
As Fisch has pointed out, these cognomina are mainly of plebeian
origin, and it is therefore not surprising that many of them which
formed gentiles have not come down to us. We may also reasona-
bly expect this to' be true, though to a less degree, in our other
groups of gentiles derived from cognomina. There is little doubt,
however, that many gentiles in -onius do not really go back to cog-
nomina in -0, but that after a large number of gentiles in -onius, from
cognomina in 0, had come into use, -onius came to be regarded as a
gentile-forming suffix and was added to any kind of cognomen.

The Origin of Roman Praenomina. 119

This tendency was helped by the fact that there frequently existed,
by the side of the cognomen in -o, a shorter one of different stem.
Thus beside Rufio, Barbo, Silo, Scribo, Aprio, Dentio, etc., there
appear Rufus, Barba, Silus, Scriba, Aj>er, Denter, etc. Several of
our names in -onius are plainly Etruscan, as Achonius, Thoceronia.
Creonius and Critonius suggest the Greek KpeW and Kpirwv. Colonius
may be referred to colonus, which is frequently used as a cognomen,
and Patronius to patronus. For Aponius we find the cognomen
Aponus, Favonius is evidently from Favonius, the west wind.
Voconius is probably to be connected with the place Voconia. Of
the other names in -onius by far the greater number are from Oscan
territory, or appear to come from plebeian Latin. A glance at the
following list will make their character quite clear :

Antonius ; Apronius (cf . cognn. Aprio, Aper) ; Aptronius (cf . CIL.
i, 81, where the name is derived from amptruare) ;x Autronius ;
Balonius (cf. cogn. Bala, and balare, to bleat) ; Caesonius, from
Kaeso ; Cauponius or Coponius, from Caupo ; Cosconius ; Qusonius,
which may be for Cosonius, an early spelling of Cossonius, from cogn.
Cosso, CIL. iii, 5542; Dexsonius (cf. Dexo, Cic. Verr. ii, 5, 42 and
gens Dexia, Gic. Fam. vii, 23, 4) ; Gargonius, found in Campania
and Picenum (cf. Garganus, a mountain in Apulia) ; Holconius,
found in Lucania (cf. holcus, Gk. oAkos, a kind of grain); Hordionius
or Hortionius, from hordeum, barley ; Laronius ; Numistronius, from
Numistro, a town in Lucania ; Numonius or Nummonius, from cogn.
Nummus ; Paconius ; Petronius, from petro, a rustic ; Pomponius, per-
haps from pompa;2 Ragonius (cf. raga, a form of bracd); Salonius,
irom*salo 'salt-dealer,' from sal; Scalponius, from scalpere, to en-
grave ; Scribonius, from scribo, a writer ; Sempronius ; Socconius, from

1 Deecke, Etrusk. Forsch. u. Stud, iii, 60, attempts to derive this name from
the Etruscan, citing the Etruscan name Apatru.

2 We find great difficulty in deriving Pomponius and Pompeius from an Oscan
numeral *pompe 'five.' We should expect the forms to be made from the
ordinal *pompts, in which case the -t- would remain, as we actually find in Pontius,
Lat. Quintius. On the other hand, we hesitate to derive these names from a Gk.
word pompa. An Italic derivative *pompo ' one who takes part in a procession,'
a very proper word for a cognomen, is hard though perhaps not impossible ; cf.
Comatus, Lamas, etc.

120 George Davis Chase.

soccus, a slipper, connected with the stage of comedy ; Trebonius or
Terebonius (the latter form, with e developed before the r, & process
regular in Oscan but extremely rare in Latin, shows the name to be
probably Oscan ; that it is not, however, to be connected with the
Oscan triibom, a building, is proved by the short quantity of the e in
Hor. Sat. i, 4, 114, Trebdnt) ; Teltonius ; Vatronius, from vatrax,
' having crooked feet ' ; Vennonius, from venno ' one who rides in a
benna ' (cf. Paul, ex Fest., p. 32) ; Villonius, from villus ' shaggy hair.'
The ending -eius is used to form a large number of gentile names.
They are of most frequent occurrence in Oscan territory and are
often spelled -aius, as Anaius, Popaius (for Pompeius), Ulaius, Vir-
riaius, Vibidaius. Inasmuch as the root part, also, of most of these
names is Oscan, we conclude that it was a regular Oscan formation.
Praenomina in -as, which were -a stems, were common in Oscan, as
Cahas (Zvetaieff,1 232), Tanas {ibid. 102), Voltai (ibid. 72 b), a
Faliscan genitive singular. These formed gentiles in -aius in Oscan,
and -eius in Faliscan. Thus from Maras (Indg. Forsch. ii, p. 435, vi,
1. 8) we find the Oscan forms Marahis (ibid, iii, 1. 6), Marahieis (ibid.
v, 1. 1), and Marah . . , . . rahiis (ibid, vi, 11. 4, 6, 8), Maraies (Zvet.
249), Maraiieis (ibid. 95), while the Faliscan form Mareio appears in
Zvet. 76. So also Pompaiians (Zvet. 143) is in Latin Pompeianus.
In Latin these gentHes were oftenest written -eius, but sometimes
evidently -ius, since we have Coccius beside Cocceius, Luccius beside
Lucceius, Cicerius beside Cicereius, etc. Whether these actually
came to be -lus in Latin may be doubted, since such names as
Marius may not stand for Oscan Maraius, but may be a separate,
shorter formation (cf. Lindsay, Lat. Lang. p. 320). There is no
doubt that the ending -eius became very much extended in its appli-
cation, and came to be applied to stems which did not end in -a ;
that is, that the whole ending -eius came to be looked upon as a gen-
tile-forming suffix. In Paelignian and Umbrian the same formation
took a somewhat different shape, — in the nominative -aes, — and
gave a number of Latin names in -aeus. Thus Paelignian Anaes is
Latin Annaeus, standing for an older Annaios. The following is the
list of gentiles in -eius in CIL. i :

1 Tnscriptiones Italiae Inferioris dialecticae, Moscow, 1886.

The Origin of Roman Praenomina. 121

Aeteius ; Afreius (cf . cognn. Afer, A/rd) ; Alleius (cf . gens Allid) ;
Anaius ; Appuleius (cf . cognn. Appulo, Apulus) ; Atteius or Ateius,
from cogn. Atta; Canuleius (cf. cogn. Kanus); Carmeius ; Cicereius
(cf . cognn. Cicero, Cicera) ; Cocceius or Coccius ; l Crepereius (cf .
Crepereia, and Varro, Z. Z., vi, 5) ; Egnatuleia (cf. Egnatia, a town
in Apulia) ; Epuleius (cf. Epulo, Verg. ^^w. xii, 459, and the fem.
cogn. Epuld) ; Farsuleius (cf . gens Farsia, farsus ' stuffed ') ; Fon-
teius ; Heioleius ; Mitreius ; Numoleius (cf . JVuma, Numitor, Numerius,
etc.) ; Modieius ; Iteius or Itius ; Laufeius ; Licculeius ; Livineius ;
Luceius or Luccius (cf. Luca, a Lucanian ; cogn. Zuaus") • Maneius
or Manneius (cf. the goddess Mand) ; Orucukius, a plebeian spell-
ing for Aurunculeius, from the city of Aurunca on the border of
Latium and Campania (cf. Zvet. 266, Aurunkud ; also gens Aurun-
ceid) ; Pabateius ; Pactumeius ; Patoleius s Popaius or Pompeius (see
Pomponius p. 119); Proculeius, from praen. Proculus ; Saufeius ;
Sueius ; Tetteius (cf. Zvet. 1, Teh's, Sabine for Tettius, and teta, a
dove) ; Tondeius, from tondeo, to shave ; Turpleius (cf. Turpilius and
cogn. Turpid) ; Ulaius ; Vargunteius (cf. cogn. Varguld) ; Vereius (cf .
vereias, Zvet. 81) ; Verguleius (cf. Vergellus, a river of Apulia) ; JV-
tuleius ; Virriaius (cf. Oscan Virriiis, Zvet. 128, for Virreius);
Vibuleius (cf. Vibius, an Oscan praenomen) ; Vibidaius (cf. Vibidis,
Zvet. 30) ; Vineius ; Vinuleius (cf. vinnulus, ' delightful ') ; Volteius,
from praenomen JW/a (Zvet. 72).

A considerable number of cognomina in -fc/#.r, particularly adjec-
tives such as Avidus, Calidus, Lepidus, Lucidus, Placidus, Umidus,
etc., formed gentiles in -M«j (see Zimmermann, in Wolflin's Archiv
f. Lat. Lex. vi, p. 270). These became so numerous that-M«.r was
looked upon as a regular gentile-forming suffix, and was used in

1 If we suppose this name to stand for Cotteius (cf. the confusion of Accius
and Attius), it may be derived from Cotta ; cf. Zvet. 239, kottcitjis, gen. sing. We
cannot, however, suppose this name to be exactly the same as Cocceius, for Lat.
-eius cannot be equivalent to the Osc. -its in such names as Viinikiis. Cf. also
CIL. x, 1 135, D. Cottius D.f. Gall. Flaccus, and 3776, M. Cottius M. f.

2 Or they may be connected with the praenomen Lucius, in Oscan Luvikis.

8 Cf. CIL. ix, 967, M. Paculeio Q.f., and the Oscan praenomen Paakul, Lat.
Paculus, Zvet. 138. The correspondence of c and t might possibly be explained
by the influence of the following /, as in poculum < * po-tlom.

122 George Davis Chase.

cases where it could have no place in the etymology of the word.
No doubt the relation of these -idius gentiles to the -idus cognomina
was lost in the popular consciousness, and Lucidius was connected
directly with Lucius or luceo ; Placidius, Vmidius, etc., with placeo,
umeo, without recognizing the intervening -idus. When the ending
-idius was applied to vowel stems a diphthong resulted, as in Ottei-
dius, Poppaedius. Whether these gentiles formed by the suffix -idius
were derived from other gentiles in -ius or from cognomina, as, for
example, whether Longidius was derived from the gentile Longius or
the cognomen Longus, is a question which will engage our attention
later. The following is the list in CIL. i, first of genuine -idius
names, secondly the larger number of analogical forms :

Ofdius, or Aufidius, from Aufidus, the river in Apulia ; Calidius,
from cogn. Caldus, for Calidus ; Lucidius, from cogn. Lucidus ;
Afiedius (cf. gens Affid) ; Aiedius (cf. gens Aia, cogn. Aiuld) ; Aledius
for Allidius (cf. gens Allid) ; Alfidius (cf. gens Alfia, in Latin Albia,
from cogn. Alius) ; Anaiedius (cf. gens Anaid) ; Atiedius (cf. gens
Attia, and Attidium, a town in Umbria) ; Novelkdius (cf. gens
Novellia and cogn. Novellus) ; Caesidius (cf. gens Caesia, from
caesius, ' steel-gray (eyed)' ; Canidius (cf. gens Cania, cogn. Kanus) ;
Clandia for * Clanidia, an Etruscan name (cf. gens Clania, of Etrus-
can origin, and Clanis, a river of Etruria) ; Considius (cf. gens
Consia, from cogn. Consus, which was also the name of a Roman
divinity) ; Crustidius, from crusta ; Epidius (cf . gens Epid) ; .Fufldius
(cf. gens Fufid) ; Hosidius (cf. cogn. Hosius) ; Longidius (cf. gens
Longia, cogn. Longus) ; Otteidius (cf. gens Otteia, cogn. £?##) ;
Pisidius (cf. cogn. /Vw) ; Pompaedius and Poppaedius (see Pompo-
nius,p. 119 ; Popidius ; Pumidius (cf. pumilio, a dwarf) ; Tamudius ;
Tetdius or Titidius (cf. Titius) ; Trutedius ; Turpidius (cf. gens -7k/--
//'//«) ; Vebidius, from Vibius, praenomen and gentile.

From diminutive cognomina in -«/«*• arose gentiles in -27mm1, as
Rufllius from rufulus, Caecilius from Caeculus, Rutilius from Rutilus,
etc., and thus arose, as in the case of -M«j, a gentile ending -*/««
which was often applied without the intermediate aid of the cogno-
men in -tilus. Since the cognomina were very frequently formed in
the double diminutive -illus, -ellus, gentiles are also frequent in -illius,
-ellius. In many cases of inscriptions, where a single / is written

The Origin of Roman Praenomina. 123

for the double, it is often difficult to tell which form is the basis.
For such names as Quintilius, Sextilius it is evident that the basis
is Quintilius, Sextulus, and not the adjectives quintllis, sextllis. We
will first give those names which seem most clearly of the -ilius type :

Aemilius, from aemulus, a rival ; Aesqullius, for Esquilius (cf.
Esquiliae, one of the hills of Rome) ; Amelius ; Ancilius (cf. cogn.
Ancus ; the Anculi were gods who ministered to other gods) ;
Aquilius (cf. cogn. Aquild) ; Baebilius (cf. gens Baebia and cogn.
Babilus) ; Bouftlius (cf. gens Bovia, or perhaps from a cogn.
pov<j>iXoi) ; Caecilius (cf. gens Caecia, cogn. Caecus ; Caeculus founded
Praeneste, ' unde putant Caecilios ortos,' Paul, ex F. p. 44) ; Caltilius
(cf . gens Caltid) ; Cartilius, evidently of Etruscan origin ; Carvilius (cf .
cogn. Carbo) ; Cloelius, Cloulius, or Cluilius (Liv. i, 50, says it was an
Alban family ; Paul, ex F. p. 55, derives it from Clolius, the companion
of Aeneas) ; Garcilius ; Gavilius (cf . gens Gavia, cogn. Gavillus) •
Hostilius (cf. gens Hostia, cogn. Hostllus, praen. Hostus) ■ Laetilius
(cf. gens Laetia, cognn. Laetus, Laetilius) ; Lavilius (cf. gens Lavid) ;
Maecilius (cf. gens Maecid) ; Magilius (cf. gens Magid) ; Mamilius
(cf. gens Mamia or Mammia, cognn. Mamus, Mammus, Mammuld) ;
Manlius for Manilius (cf. praen. Manius, fem. cogn. Manila, CIL.
viii, 5795 ; cogn. Maniius, CIL. x, 4048); Metilius (cf. gens Metia,
praen. Mettus, cogn. MeBillus, CIL. xii, 5686); Mutilius from cogn.
Mutilus (cf. Mutila, a town of Istria) ; Nomelius ; Opsilius (cf. gens
Opsid) ; Otacilius (cf. gens Otacidid) ; Bacillus (cf. gens Pacid) ;
Poplius or Pompilius (cf. gens Popidid) ; Procilius ; Quinctilius, from
praen. Quintus ; Ruftlius, from cogn. Rufus, rufulus ; Rutilius, from
cogn. Rutilus ; Sextilius (cf. gens Sextia, praen. Sextus) ; Sectilius;
Statilius (cf. praen. Statius) ; Tongilius (at Praeneste, tongere meant
noscere, Fest. p. 356) ; Turpilius (cf. cogn. Turpio); Tutulius, from
/«#«•, safe, or Osc. /iwfc, people ; Urbilius; Utilius ; Vecilius (cf.
Vecilius, a mountain in Latium ; the name, however, is found in
Etruria with the praenomen f^., which makes it almost certainly of
Etruscan origin) ; Vehilius ; Vergilius, cf. Vergiliae, and Vergellus, a
river in Apulia.

A few names in -#/*kf appear to be obtained directly from cog-
nomina in -«/«j, as Cantulius (cf. cantulus, a little song) ; Segulius ;
Tamulius (cf. fewu, a swelling in the feet, Fest. p. 360).

124 George Davis Chase.

Names in -illius, -ellius also have the appearance of being formed
directly from cognomina. It is not probable that this ever became
a movable suffix for the formation of gentiles. The list is as
follows :

Arellius ; Avillius from Aulus <* Avulus ; Caesellius, from cogn.
Caesulla;1 Camellius or Camelius, from cogn. Camillus ; Cascellius
(cf. cogn. Casca, from cascus, old-fashioned, dimin. *cascellus) ; Duil-
lius, for Duellius, from bellum ; Eppillius or Epilius, from cogn.
Epillus (cf . gens Eppid) ; Obellius, perhaps from ovillus ' belonging
to sheep ' ; Of ellius, Ofillius, or Qffilius, from cogn. Ofella, dimin. of
offa, a morsel; Petuellius ; Popillius (cf. Popellus), but also Poplius ;
Ravellius, from cogn. Ravilla, from ravulus ' rather hoarse ' ; Rupil-
lius, perhaps a mistake for Rupilius ; Rusticelius (cf. rusticellus,
clownish); Sepullius ; Visellius ; Vitellius (cf. cogn. Vitulus; vitel-
lus, a little calf) ; Petillius stands doubtless for Pettlius, from pettlus,
slender.

Cornelius might be understood as representing a class formed
from adjectives in -elis, as crudelis, but we have no instance to prove
that such a class existed. It may belong to the ilius class and be
explained as standing for * corne-elius, where by fusion a long e
results from the two short, as in prendo, from pre-kendo, etc.; cf.
Lindsay, Lat. Lang. p. 224. Cornelius would then be formed from
a cogn. *Corneus 'horny.' Cf. the cognomen Corneolus, a diminu-
tive of the same, which postulates a Corneus.

In the case of Aurelius it is noteworthy that we have the adjective
aureus and the cognomen Aureolus to compare, but it is likely that
Aurelius comes from a form older than either of these two. For a
discussion of the relation between Aurelius and Gk. 17X10?, see Cur-
tius, Gk. Etym. 401, and Paul, ex F. p. 23, ' Aureliam familiam ex
Sabinis oriundam a Sole dictam putant, quod ei publice a populo
Romano datus sit locus, in quo sacra faceret Soli ; qui ex hoc Auseli
dicebantur, ut Valesii, Papisii pro eo quod est Valerii, Papirii.' Cf.
also the Lettish heathen name AuseMs, from austs ' to grow light,'
'to dawn.'

1 Caesullae were girls with gray eyes (Fest. 274).

The Origin of Roman Praenomina. 125

A peculiar difficulty is presented by a small group of names in
-thus which are usually given -ilius in modern lexica, as, for example,
in Forcellini's Onomasticon, but for which the long -l- may be proved.
The group includes Manllius (Juv. vi, 243, Manilla), Lucllius (Juv.
i, 165 : Hor. Sat. i, 4, 6 ; 10, 2 ; ii, 1, 62), Servllius (CIL. vi, 358,
SERVlLIO, and often Serveilius in CIL. i), Acilius, Atilius, and
Venllius. Latin inscriptions show in these words a genuine long 1,
represented by simple I in the oldest inscriptions, and by the spurious
diphthong ei for a period after the time of the Gracchi. We believe
that these were made with the ending -ilius from praenomina in -ius,
but that analogy operated not in the same manner as with the pre-
ceding class in -ilius. It seems probable that the great majority in
our preceding list were formed from other already existing gentiles
as follows: from the cognomina Caecus and Caeculus were derived
the gentiles Caecius and Caecilius. Then, by a simple rule of three,
Caecius : Caecilius = Gavius : Gavilius, or as Manius (gentile): Manl-
lius i> Manlius. In the case of those in -ilius, however, the construc-
tion was always from the praenomen, and the proportion was under-
stood as follows : Hostus : Hostilius = Manius (praenomen) : *Mani-
ilius~> Manllius. So Lucllius, Manllius, and Servllius are explained
as formed by the gentile suffix -ilius directly from the praenomina
Lucius, Manius, and Servius. Whether Atilius and Acilius are merely
two forms of the same name is an open question (cf. the confusion
of Attius and Accius). In inscriptions Acilius is always written with
a single i : in Greek inscriptions the spelling varies between 'AxtXios,
'AkuXios, and 'AiceXtos. The quantity of the i appears from Juv. iv;
94, 'proximus eiusdem properabat Acilius aevi.' Atilius appears
CIL. i, 1027, as Ateilius. These names postulate a praenomen
*Acius or *Atius which is borne out by the constant spelling Atiedius,
CIL. i, 182 ; 1 167 ; ix (frequently), and Zvet. 41. Venilius appears
CIL. i, 580, in a Greek inscription, as Veneilius; elsewhere as
Venelius. No certain reliance can be placed on the Greek form;
Mommsen thinks it a mistake. We know of the name Venllia as the
mother of Turnus and also the wife of Janus. Cf. also Varr. ap.
Aug. Civ. Dei, vii, 22, ' Venllia unda est quae ad litus venit.'

From cognomina in -inus were formed gentiles in -inius, as follows:
Cacclnius, from Caeclnus ; Canlnius, from Canlnus (rare ; cf. cogn.

126 George Davis Chase.

Cams) ; Crispinius, from Crisplnus ; Fulcinius, from Fulcina; Flavi-
nius, from Flavinus ; Gabinius, from Gabinus ; Graecinius, from
Graecinus ; Latinius, from Latinus; Mesclnius ; Obinius, for Ovi-
nias (?), from ovinus ; Vatinius, cf. vatius ' bow-legged ' ; Verglnius
(cf. Sil. Ital. xiii, 824); Atinius.

A good number of gentiles were formed also in -inius from cogno-
mina in -inus, as Asinius, Geminius, etc., from Asinus, Geminus, and
from these, it seems, there arose the ending -inius, which was often
used to derive a gentile from another gentile. The following are those
in -inius of both kinds :

Asinius, from cogn. Asina ; Arsinius, Orsinius, Etruscan names (cf.
cogn. Arsinus) ; Flaminius, from cogn. Flamen; Geminius, from Gemi-
nus; Licinius, from Licinus ; Sanguinius, from sanguineus 'blood-red ' ;
Voltinius (cf . tribus Voltinid) ; Anainius (cf. gens Anaia, of Etruscan
origin) ; Aninius, Anninius (cf. gens Annid) ; Caesinius (cf. gens Caesid) ;
Cantinius (cf. gens Catttid) ; CarfiniusQ) ; Catinius (cf. gens Catid) ;
Cominius ; Fabrinius ; Fafinius (cf. gens Fafid) ; Faltinius (cf. cogn.
Falto) ; MagoZnius (cf. gens Magullia, cogn. Magula, from Magd) ;
Novercinius, from novercus; Ogulnius; Popnius for Popinius ; Safinius;
Senenius ; Sepstinius; Tamsinius (cf. gens Tampid) \ Titinius (cf. gens
27/fo, praen. Titus) ; Volminius or Voluminius, from Volumnus, the
tutelary divinity of new-born infants.

From cognomina in -<?.*•, -few, as *Cornufex, and in -£«w, as Caedicus,
Verg. ^4<?«. x, 747, there arose gentiles in -fozw, and on the analogy
of these, gentiles were formed from other gentiles by means of the
suffix -?«W. Thus we have the following :

Caedicius (cf. gens Caedia and cogn. Caedicus) ; Coricius ; Crassi-
cius(ci. gens Crassia and cogn. Crassus) ; Cornuficius ; Fabricius (cf.
gens Fabria and fabrica, a workshop) ; Fuflcius (cf. gens Fufid) ;
Limbricius ; Minicius (cf. gens Minid) ; Mundicius (cf. gens Mundia,
cogn. Mundus) ; Muncius for Municius (cf. gens Munia, cogn. Munus,
Munnus) ; Peticius (cf. .ftto, the goddess of prayer) ; Poplicius, Po-
blicius, Publicius ; Larcius, for Laricius, from /<z«w;, -&■«•; Sulpicius;
Septicius; Selicius; Vinicius ; Vespicius (cf. wj/ta 'corpse-bearer') ;

The gentiles in -«/««• have as their basis perfect passive participles
in -#».r used as cognomina, or nouns in -«■, -//«•, as follows :

The Origin of Roman Praenomina. \2J

Condetius, from Conditus ; Domitius, from Domitus ; Equitius, from
Eques ; Tarquitius (Etruscan) ; Caltius, tor Calitius (?).

Names in -atius come from cognomina in -as, atis derived from
names of places (see Mowat, Mimoires de la sociiti de linguistique de
Paris, 1868, p. 94), or from passive participles in -atus :

Baebatius ; Caf atius (Etruscan ; cf . cogn. Cafd) ; Lutatius, from
*lutatus 'muddied,' from lutum; Ocratius (cf. cognn. Ocra, Ocrea,
Ocraterus) ; Munatius ; Curiatius (cf. Curiates, a people of Umbria) ;
Egnatius, from the town Egnatia; Horatius (cf. Mowat, /. c. p. 95),
from Foretii, or Hora, the goddess ; Folcatius ; Minatius, a praenomen
in Campania (cf. Veil, ii, 16) ; Veratius.

Besides the formations which we have already listed there are
others which are less frequently found, and which perhaps never
developed movable suffixes, such as -erius < -esius, as in Valerius,
Papirius, Numerius; -ennius < -endius, from gerundives in -endus, as
Herennius, Cupiennius ; -ustius, as in Fidustius, etc. The great mass
of Roman gentiles were formed from other already existing names
by the simple ending -ius, and as such we will consider the rest of
the gentiles which we have collected. A large number of these are
certainly of non-Latin origin, and our work will be made more com-
prehensible if these are first sorted out, as many as possible, from
the others. We will first give those which both from their form and
the locality in which they are found are proved to belong to the
Oscan dialect :

Alfius, Lat. Albius, from cogn. Albus ; Ansius ; Antracius (cf.
Anthrax, ap. Plaut.) ; Arrius ; Asuius ; Babbius (cf. Zvet. 105,
B\a~]bbiis, Oscan, and cogn. Babius ; babulus, a fool) ; Blasius (cf .
cogn. Blasius, Blasio, Blaesus (?) ; Blossius ; Braccius ; Bruttius (cf .
cognn. Bruttio, Bruttius) ; Calasius (cf. Calatia, a town in Campa-
nia) ; Cluvius, Clovius, from Cluvia, a town in Samnium ; Consius,
from Consus, an ancient Italian divinity ; Eprius ; Fisius (cf. Fisus,
Fisius, an Umbrian divinity) ; Fufius ; Furrius ; Gavius, from prae-
nomen Gaius ; Grusius ; Hedius ; Heidius ; Herennius, from cogn.
Herennus < *herendus, a gerundive of the Oscan verb herio, to wish
(cf. the Lat. Cupiennius) ; Herius, from the Oscan praenomen Herius,
from herio; Macius (cf. cogn. Macus, Macio, and the Oscan name
Magius) ; Maius ; Messius (cf. cogn. Messor); Mevius ; Minius (cf.

128 George Davis Chase.

Zvet. 121, Miinieis, Oscan, and the Italian praenomen Minius) ;
Monnius (cf. cognn. Monno, Monnus, Monnulus) ; Munnius (cf. cogn.
Munnus); Nasennius, for -endius (see Herennius, p. 127); Nellius ;
Nelpius ; Niraemius ; Novius, from the Osc. praen. Nbvius ; Numi-
sius,Numpsius, for Lat. Numerins ; Occius ; Orcevius, Orcuius ; Orfius
for Lat. Orbius (cf. cogn. Or/a) ; Pacius, Paquius, from the Osc.
praen. Pacius ; Pandius (cf. the goddess Panda) ; Papius (cf. cogn.
Papilus) ; Patlacius ; Pettius ; Pontius, for Lat. Quinctius (Pontus <
*Pomptus'=\AX.. Quintus); Pulius, Pullius (cf. pullus, dark-colored);
Raecius ; Pains ; Ronius ; Rnntius ; Sadrius ; Satrius ; Salvius, from
Osc. praen. Salvius; Stlaccius ; Telepius ; Taracius ; Toutius, Tutius,
from Osc. tovto, people ; Trosius ; Tuccius ; Turius ; Urnannius, from
* Urnannus, from an Osc. verb *urnaum; Utius ; Vedius; Veserius,
from Veseris, a town in Campania ; Vesvius (cf. Vesuvius, the vol-
cano in Campania ; Vesulliais, Zvet. 93 ; Vesulias, an Osc. goddess,
Zvet. in, b) ; Vibius, from Osc. praen. Vibius ; Vinucius (cf. Viinikiis,
Zvet. 143); Vitrovius ; Volusius, from praen. Volusus ; Vulius; Pe-
scennius, Percennius, Pessennius, from Osc. *persccnnus (Lat. precandus).

A large number of names are certainly not of Roman origin, but
belong to some of the other Italic dialects. We will class together
those which are evidently from Umbria, Picenum, the Sabines, and
the southern dialects of Latium, distinguishing where it is possible :

Aburius ; Aebutius ; Aigius, Volscian ; Ampius ; Andius, Picenum ;
Annavius, Annius, Picenum; Aprufenius ,-1 Audasius; Aufustius^ (cf.
gens Aufia, Aufillid) ; Babrius, Picenum ; Camurius, from camur
' turned inward ' (of the toes) ; Cincius, or Cintius; Cossutius (cf. the
Volscian name Cossuties, Zvet. 47 ; cognn. Cossus, Cosso, Cossinus, and
gentt. Cossius, Cosseius, Cossicius, Cossitius, Cossinius, Cossidenius, and
Cossonius) ; Crepusius, from Crepuscus < born at twilight ' (cf . creper,
Varr. L. L. vi, 5) ; Decius, an old Ital. praenomen (cf. Liv. xxiii, 7,
10, Decius Magius) ; Faderius (cf. gens Fadia, Fadena, cogn. Fadus) ;
Fuulius, Umbrian (ci.fuligo, soot) ; Geganius, an Alban family, Liv.

1 From *Apro-fen, boar-killer. Cf. Gk. ipbvos, Skt. {g)kan, and see Brugmann,
Vergt. Gram, i, 319. The word appears to show a very interesting survival of
an Indo-European name.

2 The same ending -ustius appears as in Fidustius, q. v.

The Origin of Roman Praenomina. 129

1, 30; Gellius ; Gessius ; Liguius; Lucretius (cf. Lucretilis, a moun-
tain among the Sabines) ; Maecius, Picenum (cf. cogn. Maecinus) ;
Malms, Mallius (see Mommsen, ap. Borghesi, CEuvr. ii, 219, n.) ;
Mimesius, Picenum ; Mindius ; Modius ; Nerius, Umbrian (cf . cogn.
Nerd) ; Obulcius (cf. Obulco, a town in Spain) ; Oppius ; Opisius,
Sabine (cf. cognn. Opera, Opita, praen. Opiter) ; Orcunius ; Ovius,
Paelignian (cf. Zvet. 21 and 27, and the Oscan praenomen Ovius);
Pilius, Peilius ; Poinisius, Picenum; Rustius ; Sangurius, Picenum;
Seppius ; Staberius ; Tadius ; Teidius ; Tettienius ; Tetius ; Tillius ;
Titius, Sabine, from praen. Titus ; Tampius (cf. cogn. Tampilus) ;
Tattius, Sabine.

The following are Etruscan :

Abelesius ; Ancharius (cf. the goddess Ancharia) ; Arnustius, from
Arnus, a river in Etruria ; Arruntius, Etruscan Arruns ; Clanius,
from Clanis, a river of Etruria; Larthius, from Etruscan Lars ; Otius ;
Proenius ; Sudernius ; Tanius, Thanius ; Telutius ; Veisinnius ; Vel-
tius ; Vensius ; Vessius ; Vbkacius, Volchacius.

From Genua we have the two native names Meticanius and Pelia-
nius.

Of gentiles which are Roman, or which are found at Rome and
cannot be shown to have originated elsewhere, there are the following :

Aelius ; Agrius, from ager (cf. cogn. Agellus) ; Albius, from
cogn. A/bus; Allius, from the old Latin name Alius (cf. Deecke,
Etrusk. Forsch. u. Stud, v, p. 104) ; Antestius, Atistius, from antistes,
a priest; Antius, from Antius, the adjective from Antium; Appius,
from praen. Appius, originally Sabine ; Aprius, from cogn. Aper ;
Aqutius, for Acutius, from cogn. Acutus ; Artius from ater, black (Varr.
L. L. viii, 80 ; x, 44) ; Attius, from Sabine Atta, Attus ; Aulius, from
praen. Aulus; Avius, from avus (cf. cognn. Avitus, Aulus < * Avulus,
Aviold) ; Axsius, from * axa (cf. cogn. Axilla) ; Baebius (cf. cogn.
Babilus V) ; Caeicius, for Caecius, from cogn. Caecus ; Caedius, from
caedo; Caelius, cf. Caelius mons, said to be named after the Tuscan
Caeles Vibenna ; Caesius, from cogn. Caesius 'bluish-gray-eyed'; Caius,
from the praen. Gaius ; Calpurnius (cf. Paul, ex F. p. 36, 'a Calpo,
Numae regis filio ') ; Calvius, from cogn. Calvus ; Casius, Cassius,
from cogn. Cassus ' empty,' older Casseius > Cassius ~> Cassius
(cf. Ritschl, de Sepulcro Pur.) ; Cqtius, from cogn. Catus (cf. Cato

130 George Davis Chase.

and Catulus) ; Caucius (cf. caucus, a drinking vessel) ; Cauca, a
city of Spain) ; Cervius, from cervus (cf. cogn. Cervid) ; Cestius,
from cogn. Cestus ; Cipius, from Cipus, a fabled Roman praetor ;
Cispius (cf. Cispius mons, one of the peaks of the Esquiline) ; Clau-
dius, Clodius, from cogn. Claudus ; Coilius, Coelius, Caelius (cf . Caelius
mons) ; Comicius, from cogn. Comicus (?) ; Cordius, from cogn.
Cordus ; Coriarius, from coriarius, a tanner ; Cossius, from cogn.
Cossus ; Cottius, from cogn. Cl?## / Cupiennius, from cupiendus (cf .
Osc. Herennius) ; Cupius (cf. cognn. Cupio, Cupitus) ; Curius, from
cognn. Curius, Curio ; Curtius, from curtus, short ; Decumius, from
praen. Decimus ; Deidius, from cogn. Z?z'</(Z (?) ; Dindius (cf. cogn.
Dindid) ; Drusius, from cogn. Drusus ; Gallius, from cogn. Gallus ;
Granius (cf . granum, grain ?) ; Faberius, from cogn. Faber ; Fabius
(cf. cogn. Fabatus, from /aba) ; Fannius, for *Fadinius~>*Fadnius~>
Fannius (cf. cogn. Fadus, and gentes Fadia, Fadonia, Fadiend) ;
Fidustius, from Jidusta, ' a fide denominata ea quae maximae fidei
erant ' (Paul. «* .# p. 89 *) ; Flavius, from cogn. Flavus ; Fulvius, from
cogn. Fulvus ; Furius, for older Fusius, from cogn. Fusus (cf. -4fr.
Furius Fusus, cos. 308, B.C.) ; Furnius, trom/urnus, oven (?); Helvius,
from cogn. Helva ; Hirtius, from hirtus, shaggy ; Hortensius, from
cogn. Hortensis, from hortus (?) ; Hostius, from praen. Hostus ; Iulius,
from lulus, seep. 143 ; Junius, for * luvenius, from cogn. Juvenis ;
luventius, from cogn. Tuventus (cf. iuventa) ; Laelius, perhaps for
Lavilius, from the gens Lavia (cf . caelum < * cavilum) ; Laetorius,
from * laetor (cf . cogn. Laetus) ; Libertius, from Libertus, a servile
name (cf. cognn. Libertio, Libertinus) ; Ligurius, from cogn. Ligur ;
Llvius (cf. Itvidus (?), bluish, from /«»«>) ; Lollius, from cogn. Lolla,
from cogn. Laurus (?) (*Zaurulus > * Laullus > Lollus) ; Lucius, from
praen. Lucius ;" Lurius, from cogn. Lurus (cf. luridus, sallow) ;
Luscius, from cogn. Luscus ; Maenius (cf. maena (?), a kind of sea

1 Pauli, Altital. Stud, ii, 140, thinks that Au/ustius, Fidustius, Rustius, etc.,
show IEur. -z>Aw / cf . Skt. vasistha, frestha, etc. Gk. 4/w<rros, KiXXwrros, Kpirurros,
etc.

3 Gentiles derived from praenn. in -z'«j are identical in form, as Lucius, Servius,
Manius, etc. It is a question whether the forms were transferred without any
further formative suffix, or whether the Latin gentiles Lucius, etc., stand for an
earlier Lucius, etc. See the discussion of Marius, Marcius, p. 1 20.

The Origin of Roman Praenomina. 131

fish) ; Magnius, from cogn. Magnus; Mamius, Mammius, from cogn.
Mammus, Mammulus ; Manius, from praen. Manius ; Marcius, from
praen. Marcus ; Marius (cf. cogn. Maro, Osc. praen. Mara); Memmius,
from cogn. Memmus ; Minucius ; Mispius; Mucius (cf. miicidus (?),
snivelling) ; Mulvius ; Mummius, from cogn. Mumtna ; Mustius,
from cogn. Mustus ; Naevius, from praen. Gnaeus (?) ; Nonius, from
nonus (cf . Nona, a fem. name) ; Numerius, from praen. Numerius ;
Numitorius, from cogn. Numitor; Nummius,ixom cogn. Nummus; Octa-
vius, from cogn. Octavus ; Opimius, from opimus ; Oratius, from cogn.
Orata ; 1 Orbius, from cogn. Orbus ; Papirius ; Plnarius ; Placentius,
from placens, charming ; Plaetorius, Pleturius ; Plancius, from cogn.
Plancus ; Plautius, Plotius, from cogn. Plautus ; Plutius, from
Plutus (?) ; Pollius, from cogn. Pollus, for Paullus (but cf. also cogn.
Pollid) ; Porcius, from porcus ; Postumius, from cogn. Postumus ;
Propertius, an Umbrian name (cf. Buecheler, Umbrica, p. 172, £W.r.
iter. Propartie, and Prop. iv. i, 125); Prosius (cf. .P/wa (?), the
goddess presiding over births with the head foremost) ; Pupius, from
pupus, a child ; Quinctius, from praen. Quinctus ; Ramnius (cf. Ramnes);
Rancius, Rantius (cf. rancidus, rank smelling) ; Remius, from cogn.
Remus ; Renius (cf. r^««?j ?) ; Rocius, from cogn. Rocus ; Roscius (cf.
roscidus (?), dewy) ; Rubius ; Rubrius, from ruber; Rufius, from
cogn. Rufus ; Rullius, from cogn. Rullus ; Saleivius, from saliva (?)
(cf. Mucius); Saltorius, from * Saltor, 'dancer'; Secius, from cogn.
&«« (cf. Wolflin, Archivf. Lat. Lex. vi, p. 269); 5«'«j, from ,S«<z,
the tutelary divinity of sowing ; Sentius, from xot/w (?), thorny ;
Septitnius, from cpgn. Septimus; Sergius, from Servius (?) (cf.
Vanicek, Etym. Worterb. p. 1026); Sertorius, from praen. Sertor ;
Servius, from praen. Servius ; Sextius, from praen. Sextus ; Silius,
from cogn. Si/us; Sirtius ; Specius ; Spurius, from praen. Spurius;
Statius, from cogn. and praen. Statius ; Statorius, from Stator, epithet
of Mars and Juppiter ; Sucius, from jz^w.r (?), vigor ; Suetius, from
suetus; Taurius, from cogn. Taurus ; Terentius, from teraw, 'grinder,'
' thresher ' ; Thorius, from few, muscle, brawn ; Tiburtius, from

1 Cf. Varr. .ff..ff. iii, 3, 10, sic nostra aetas in quam luxuriam propagavit lepora-
ria hac piscinas protulit ad mare et in eas pelagios greges piscium revocavit. Non
propter has appellati Sergius Orata et Licinius Murena ?

132 George Davis Chase.

Tiburtes ; Titanius, from Titanus ; Titurius; Tossius, for Tonsius (?),
from tonsus, shaved ; Tullius, from praen. Tullus ; Ursius, from
ursus, bear; Valerius, older Valesius (cf. Zvet. 30, Paelig., Valesies),
from Volesus, praen. (cf. Vblero) ; Valvius ; Varius, from cogn.
Varus ; Veius, from Veius, adjective from Veii ; Vestorius ; Vettius
(cf. cogn. Vettd) ; Veturius, from vetus ; Vicensumarius, from the same
word, a receiver of the tax of the twentieth part ; Vitorius, for Victo-
rius (?), from cogn. Victor ; Vilius, from vilis (?), trifling ; Vivius,
from vivus.

The results of our examination of gentile names may be briefly
stated as follows : all gentiles, so far as can be proved, are adjectives
derived from words which were either praenomina, cognomina, earlier
gentiles, or words of the kind that were used as cognomina. This
is shown to a degree which must appear very conclusive. We be-
lieve we have proved that all names which ended in -anus, -anius,
-ienus, -onius, -eius, -Idius, -ilius, -llius, -ellius, -elius, -inius, -Inius, -icius,
and -alius were of this kind. We have furthermore shown that a
large number of names found in Latin sources were certainly not of
Roman origin, and so do not directly bear on our investigation. Of
those which are Roman, or cannot be proved otherwise, and which
end simply in -ius, we have about the following proportion :

From known cognomina 57

From known praenomina 19

Having a common derivation with known cognomina 13

From words which might have been used as cognomina 47

Of unknown origin 13

Total 149

Of the few unknown there is nothing in the form or appearance to
mark them as in any way different from the rest ; most of them
probably belong to non-Latin dialects or are forms worn down by
long usage into unrecognizable shapes. Our results must convince
any one that gentilicia could not have existed before the praenomina
and cognomina, from which they are derived, were already in exist-
ence, and that they therefore are the representatives of no original
Indo-European name system. It must be, therefore, since we have

The Origin of Roman Praenomina. 133

excluded cognomina and gentilicia, that in praenomina, if anywhere,
are to be found the traces of Indo-European names of which we are
in search.

V. Praenomina.

The subject of Roman praenomina early attracted the attention
of antiquarians and philologists. Varro is known to have written
a separate work on the subject and has also left us many items
of interest in his treatise de lingua Latina. The epitome of an
anonymous liber de praenominibus appended to the work of Val-
erius Maximus is also of extreme interest, as are numerous ref-
erences in Quintilian and the later grammarians. The Roman
writers give us some idea of what among them were considered
praenomina ; but in attempting etymologies they are often obviously
so wide of the mark that we cannot but believe that their knowledge
of the true origin and distinction of praenomina was at least as
unreliable as our own. Their tendency is to refer the meanings of
praenomina to circumstances connected with the birth of the child.
In certain names, as Vopiscus, Spurius, Postumus, Quintus, this is
obviously correct, and it is probable that their analogical inclina-
tions, perhaps often aided by a deceptive Volksetymologie, led them
to attribute similar origins to such names as Marcus, Lucius, Manius.
We are bound to bear in mind the conclusions of the ancients and to
treat them with the respect due to their antiquity and to the evident
honesty with which they were sought ; but we are under no con-
straint to accept them, or to believe them obtained in the light of
greater knowledge than our own. The advantages of comparative
study which we possess, added to the accumulations of centuries,
make us masters of the subject in hand, as well as of the funda-
mental principles of linguistic development, to a degree not attained
by Varro and the other Roman antiquarians.

Among modern writers the subject has made little advance.
Hubner (Quaestiones Onomatologicae Latinae, 1854), Mommsen
{Romische Forschungen, Berlin, 1864), and Schneider {Beitrdge zur
Kenntniss der Romischen Personennameti) have pursued the investiga-
tion on the lines of Varro, and have done little more than follow in
his illustrious footsteps. It is true they have succeeded in gathering

134 George Davis Chase.

many facts and items of value and interest ; but they have arrived
at no general or systematic results which could for a moment be
admitted as conclusive. Mommsen has gone farthest and has pretty
well established the extent of the use of praenomina among the
patrician families at Rome. For a review of the literature of the
subject and also for examples from Latin literature of some of
the rarer praenomina I am indebted to a manuscript dissertation
of Dr. W. T. Piper, deposited in the Harvard University library.

While the literature of the subject of praenomina contains much
of value, it also contains errors. It is not our present purpose, how-
ever, to review and discuss the theories of previous writers, but
rather, pursuing our own way, to attempt to reach an adequate
explanation of Latin praenomina in their development and as a
system. To accomplish this we shall follow the lines we have laid
out in the preceding chapters on cognomina and gentilicia, except
that in the present chapter, instead of restricting ourselves to a
portion of the field, we shall take all the names that are even
suspected of being praenomina, and try to find out the story which
they themselves tell.

Before we take up the names themselves let us recall a few perti-
nent facts about praenomina. While the gentilicium was the general
designation to which a person was born and which he perforce
assumed, derived, as we have seen, from earlier names ; and the
cognomen was a name given to the individual or his Roman ancestor
as in some way personally descriptive ; the praenomen was the name
arbitrarily selected from a limited number, with no other restriction
as to choice than that it had perhaps been commonly held among
the members of his family. This fact of arbitrary selection is an
important one in our discussion ; for it marks the first characteristic
common to the Latin praenomen and the traditional Indo-European
name, — a characteristic which at the same time excludes the gentile
and the cognomen from a possible origin outside of Italic bounds.
To a Roman the formal sign of a praenomen was its position before
the gentile. So universally was this the rule that we find no sure
instance in republican times of the praenomen placed after other
names, except occasionally in verse, where the exigencies of metre
constrained the poet, as in the Scipio epitaphs, CIL. i, 29 and 33.

The Origin of Roman Praenomina. 135

The name Alfenos Luci, in CIL. i, 831, is of doubtful interpretation ;
Luci may be a genitive. As in most discussions of praenomina cer-
tain ones rarely met with in Latin play a prominent part, it is
desirable at the outset to correct any erroneous impressions about
the comparative extent to which these names were actually used.
Perhaps in the first volume of CIL. a greater variety of praenomina
is found than in any equal space, and the following figures, giv-
ing the number of occurrences of the several names in that volume,
will accordingly afford a fairly correct idea of the relative fre-
quency of their use : Lucius 535, Gains 527, Marcus 404, Quintus
241, Publius 224, Gnaeus no, Aulus 99, Titus 82, Sextus^j, Manius
30, Numerius 27, Decimus 22, Servius 19, Tiberius 17, Spurius 16,
Appius 10, Vibius 14, Postumus 3, Novius 3, Statius 9, Plautus 2,
Salvius 7, Pacius 4, Epius 2, Vel. (?) 3, Sertor 2, Paulus 2, Trebius 2,
Nero 3, Marius, Tullus, Pup., Otanius, Lar., Ar., Ovius, Mesius, Vo.,
Va., Opi., Kaeso, Herius, 1 each.

Add to this list a number of names from Livy, the Incertus Auctor
de praenominibus, and inscriptions, all of the rarest occurrence, —
Volusus, Voter 0, Numa, Pet., Gellius, Ancus, Vopiscus, Proculus,
Geminus, Taurus, Cossus, Caesar, Agrippa, Minius, Minatius, Cordus,
lulus, Hostus, Septimius, Paustus, Mamercus, and Denter, — and we
have the largest list that could be claimed as praenomina, — a total
of sixty-four. In fact no one writer claims so many, but since each
of the sixty-four has by one or another authority been accredited
with being a Roman praenomen, we must examine the right of each
one to be so ranked.

Many of the names on this list are native in sections of Italy out-
side of Latium, and oftenest appear in inscriptions from provincial
localities, but were always liable to be carried to Rome. The
Etruscans are, according to Deecke and Pauli {Etruskische Forsch-
ungen und Studieti) responsible for a large number. They certainly
furnished the following:

Aruns, very frequent in Etruscan inscriptions in the form Arnth.
It appears in the name Ar. Sudernius in CIL. i, 1363, from Etruria.
Cf. also Dion. Hal. iii, 46, TvpprjviKa Oefievos auTots ovo/xaTa, t<3 Aiev
Apovra, rai Si AvKOfiwva.

Lar, very common in Etruscan inscriptions in the form Larth.

136 George Davis Chase.

Incert. Auct. 4, Lartis praenomen sumptum est a laribus, Tuscum
autem esse creditum, fuitque consul Lar Herminius cum T. Verginio
Tricosto. So also Dion. Hal. xi, 51 ; Diod. xii, 27, 1. But Livy,
iii, 65, gives the name as Sp. Herminius. The name Lar occurs in
Lar Ancharia, from Etruria, CIL. i, 1371. Cf. Dion. Hal. v, 21,
Ba<riA.evs KAovcriaiw r<ov Iv Tvpptjvia Aapos ovofw. Uop&ivos iiriK\y)<rtv.

Otanius occurs only in CIL. i, 1395, in Otanius Larthius, from
Etruria.

Vel., an abbreviated praenomen, occurring in inscriptions from
Etruria in CIL. i, 1345, 1359, 1360, 1377.

Vo., read by Mommsen in CIL. i, 13 13, in L-VECILIOVOF-,
and interpreted by him as Volusi filius, but differently read by others.

A considerably larger number of our list can be proved to be
purely Oscan praenomina. Their occurrence is mainly in Oscan
inscriptions, or in Latin inscriptions from Oscan territory, or from
other quarters of Italy in names which are evidently Oscan. They
are only occasionally found in Rome or Latium. In Latin they
nearly all have the ending -ius, which in Oscan is written -is. From
these were freely derived gentilicia in Oscan in -Us (-tyo-) ; cf. Lind-
say, Lat. Lang., p. 32 b. Thus from praenomen Statis, in Latin
Statins, from the stem stat-io-, is derived the gentile Statiis, from the
stem stat-iyo-, which in Latin is also Statius. In Latin the two forms
have been reduced to one, in -ius. Thus we have, in the case of
Oscan names in Latin, seemingly the same name used equally as
praenomen and gentile. It is only when we turn to the Oscan mon-
uments that we perceive that one is a derivative of the other (see
above, the gentile Lucius, p. 130). The following are the praeno-
mina in our list that are certainly Oscan:

Vibius. Liv. xxvii, 15, et Bruttiis similis spes veniae facta est, cum
ab eis Vibius et Paccius fratres longe nobilissimi gentis eius, etc.
Ibid, xxiii, 6, Vibius Virius, one of the Campanian ambassadors (cf.
Zvet. 128, the Oscan gentile Virriiis). Liv. xxv, 14, Vibius Accaeus,
prefect of the Paelignian cohort. Zvet. 126-7, Vibis Smintiis; 83,
ViibisOhtavis ; 130, Vibis ; 1293, Vibiiai ; 129™, Vibiiai Akviiai. Rhein.
Mus. xlv, p. 162, Opfl. Vi. Pak. Tantrnnaiom .... In CIL. ix and
x, containing the inscriptions from Oscan territory, Vibius, occurs as
gentile 230 times. In Latin inscriptions we find the following :

The Origin of Roman Praenomina.

137

IL

, i, 1274.

V. Blasius.

Samnium.

ii

" 1249.

V. Popidius.

Pompeii.

ii

" 1181.

V. Ofellius.

Volsci.

ii

" 625.

V. Vibidaius.

Marsi.

ii

" 182.

V. Atiedius.

tt

ii

" iS4ia.

V. Salvius.

it

it

" 1412.

V. Solsienus.

Umbria.

ii

" 1286.

V. Novelledius.

Paeligni.

ii

" 1285-

V. Aienus.

<<

li

" 1279.

V. Pettius.

tt

ii

" 75-

V. Anicius.

Latium.

U

" 1097.

V. Vedius.

Rome.

a

" 1456-

V. [ ]etius.

Aquileia.

a

" P-555-

V. Alpis.

(unknown.)

Fast. Cons. 302 a. U. c, P. Sestius Q. f. Vibi n. Capito [Va]ticanus.

Ephem. Epig. i, 138, L. Villius V. f. ; also ibid. 38, 120, 123, and
elsewhere.

Statius. Liv. ix, 44, Statius Gellius imperator Samnitium ; xxiv,
19, Q. Fabius consul ad Casilinum castra habebat, quod duum
milium Campanorum et septingentorum militum Hannibalis tene-
batur praesidio. Praeerat Statius Metius missus ab Cn. Magio
Atellano, qui eo anno medix tuticus erat ; xxiii, 1, accitus in Hirpi-
nos a Statio Trebio pollicente se Compsam (in Samnium) traditurum.
Compsanus erat Trebius, nobilis inter suos.

Zvet. 98. C (?) Staatiis L. Klar; 128, Statiis Gaviis ; 230, T.
Statiis ; 109, Statis Cloil. C. ; 233, CTATIC . . . . ; 132, Statie.
Indo-Germ. Forsch. ii, 1893, p. 437, ii, 1. 9, Stat ....

IL

. i, 1266.

Sta. Raius.

Apulia.

a

" 184.

Sta. Fl[avius].

Marsi.

a

" 183.

St. Magius.

M

a

" 103.

Sta. Cupius.

Latium.

it

" 1292.

St. Pomponius.

Amiternum.

a

" 169.

Sta. Tetius.

Picenum.

a

" 1460.

St. Mulvius.

Aquileia.

a

" p. 142.

St. Voc[ ].

Spain.

a

" P- 555-

St. Pontius.

(unknown.)

138 George Davis Chase.

These examples, particularly those from Livy, of distinguished
Samnite leaders, seem at first sight not at all in accord with the
statement of Aulus Gellius, iv, 20, 12 : Statius autem servile nomen
fuit. Plerique apud veteres. servi eo nomine fuerunt. Caecilius
quoque ille comoediarum poeta inclutus servus fuit, et propterea
nomen habuit Statius. Sed postea versum est quasi in cognomen-
tum appellatusque est Caecilius Statius. From our earlier evidence
we are forced to believe that in origin Statius was a common and
creditable Oscan praenomen. At Rome it might have become
appropriated by the servile class and so have fallen into disuse
among free-born people. This is the easier to believe since it was
native, not at Rome, but in the provinces. In earlier times many
slaves came from the Oscan states, and the Oscan name was always
either openly or secretly regarded with hatred and contempt at
Rome.

Numerius is an Oscan praenomen actually found in Oscan inscrip-
tions. Cf. Zvet. 137, [Nijumsis Heirennis and Niumsieis ka . . . . ;
252, [Map]as IIo/u,7rTies NivprSojis ; and as a gentile, 102, Tanas
Niumeriis. Its introduction to Rome is explained in Festus, p. 170:
Numerius praenomen numquam ante fuisse in patricia familia dicitur
quam vis * Fabius qui unus post sex et trecentos ab Etruscis inter-
fectos superfuit, inductis* magnitudine divitiarum uxorem duxit
Otacili Maleventani, ut turn dicebantur, filia * ea condicione ut qui
primus natus esset praenomine avi materni Numerius appellaretur.
So also Incert. Auct. 6, Numeriis sola tantummodo patricia familia
usa est Fabia, idcirco quod trecentis sex apud Cremeram flumen
caesis, qui unus ex ea stirpe extiterat ducta in matrimonium uxore filia
Numerii Otacilii Maleventani sub eo pacto ut quem primum sustulis-
set, ei materni avi praenomen imponeret, obtemperavit. Of the
twenty-seven examples of Numerius as praenomen in CIL. i, twelve
come from Campania and Lucania.

Novius. Zvet. 250, NO-COMNI/NO, Oscan ; 19, VOBELIESNO,
Paelignian. For the very frequent occurrence on Oscan soil of
Novius as a gentile, see CIL. x, p. 1048.

CIL. i, 1 261. Nov. Vibius. Lucania.

" " 96. Nov. Comenius (?) Latium.
" " 878. Novi Graeci. "

The Origin of Roman Praenomina. 139

Pacius or Paquius appears to be a purely Oscan praenomen, par-
ticularly frequent in Lucania. Liv. xxvii, 15, Vibius et Paccius fratres
longe nobilissimi Bruttiis. Zvet. 86, Pakis Tintiriis ; 1292, Pakiu
Kluvatiud ; i2910, Pakim ; 1299, Pakis ; 202, .... pask Pak ; 89, Pk.
De. Pk. sovad eftiv upsed, ' Pacius Decius, the son of Pacius, built
with his own money'; 104, ... . | t. Pk. Laf. Pk. ; 236, A. Aairows
TlaKprjis, in Latin, A. Laponius Paquii ; 35, CIA-PACIA MINERVA
(gentile), Paelignian ; see Zvet. Appendix, the conjecture of Deecke
that C(A)IA is meant ; 43, PA-VI-PACVIES, Marsi; Rhein. Mus.xlv,
p. 162, Opfl. Vi. Pak. Tantrnnaiom . . . . ; Indogerm. Forsch. ii,
P- 437. i» L 5> Pak

CIL. i, 1257. Paq. Aesqullius. Lucania.

" " 1262. Pac. Vitellius. "

" " 1542. Paq. Scalponius. "

" " 183. Pac. Anaiedius. Marsi.

As an Oscan gentile, Pacius or Paccius appears in Liv. x, 38, Ovio
Paccio, and very frequently in CIL. x.

Trebius. Zvet. 246, Tpe/fcs 2. Stores oeoer ; 149, Ni. Trebiis Tr.
med. tov. ; 252, TR- PLATORIVS- TR ; 128, Tr. Flapiu (?) Vfrn'iis;
93, Nv. Vesulliais Tr. m. t. ; 34, TR-AMDIS- In an inscription from
Capua lately discovered {Indogerm. Forsch. iv, p. 259) appears the
name Tr. Vfrriiefs Kenssurinefs. CIL. i, 1257, L. Cai Tr. f., Lucania.
Trebius also occurs as a gentile name not infrequently in CIL. x,
and in Zvet. 149, 198, 200. The name Trebius is probably to be
connected with the Umbrian divinity Trebus lovius.

Marius. Liv. xxii, 42, Marium Statilium cum turma Lucana ; xxiii,
7, Marium Blosium praetorem Campanum. Veil. Pater, ii, 16,
Marius Egnatius, one of the Italian leaders in the Servile War. CIL.
i, 1232, C. Pontius Mari f., from Samnium. In Oscan we find not
Marius but Mara, a common praenomen. It looks as if the Romans
in using the name changed its Oscan form Mara to Marius to make
it more in keeping with the other Oscan praenomina Pacius, Novius,
Ovius, etc.

Ovius. Liv. ix, 7, Ofillius Calavius Ovii Alius, of Capua ; ibid. 26,
Calavios Ovium Novium, of Capua ; x, 38, Ovio Paccio, a Samnite ;

140 George Davis Chase.

Zvet. 248, Ov. Afaries Ov. ; 252, Ov. Caisidis Ov. ; 21, Saluta Obel.
Ov. ; 26, Min. Rufries Ov. 1.; CIL. i, 1265, Q. Ovius Ov. f., from
Apulia.

Mesius. CIL. i, 1275, Helviae Mesi f., Samnium. As Mesius
does not elsewhere occur as a praenomen we cannot be certain that
it was regularly so used in any section of Italy.

Herius. Li v. xxiii, 43, Herius Pettius, Nolanus ; Veil. Pater, ii,
16, Herius Asinus dux Italorum ; CIL. i, 62, C. Placentius Her. f.,
from Latium. The name is evidently connected with the Oscan verb
herio, volo, and the goddess Herentas who corresponded to the Latin
Venus. The longer form Herennus < * Herendus (cf. Lat. Cupkn-
nius, CIL. i, 105 1) occurs as an Oscan praenomen in Zvet. 225.

Minius. Liv. xxxix, 13, Minium et Herennium Cerrinios, Cam-
panians. In Cato, R. R. 151, Keil following Victorius reads Manius
Percennius Nolanus. Codd. AR., however, read Minius, and V
reads Memius. Manius we should expect to find abbreviated ;
Memius we do not elsewhere find as a praenomen. Minius, from
the substantiation of the other cases which we cite, appears to be a
frequent Oscan praenomen and must have been common at Nola.
We should prefer, therefore, to read Minius in our text of Cato.
Zvet. 284, Mi. Iefis ML; 251, C. Soies Min.; 124, Min. V . . . ;
112, Miniefs Kafsilliefs Minatefs; 122, Upfals Salaviis Minies; 26,
Min. Rufries Ov. 1. ; from Capua (Indog. Forsch. iv, p. 259), Mi.
Blossii ML; CIL. i, 1230, M-MAGI-MIN-F-SVRVS, Apulia; Rhein.
Mus. xliii, p. 128, Mi. Anniief(s).

Minatius or Minatus. Veil. Pater, ii, 16, Minatius Magius, of
Apulia. In Oscan the name has the form Minatus (cf. the change
of Oscan Mara to Marius in Latin). Zvet. 112, Miniefs Kafsilliefs
Minatefs; Rhein. Mus. xlv, p. 162, Mina. Naseni ; Indg. Forsch. ii,
p. 437, ii, 1. 7, Minaz = Minats < Minatus.

Opius. Zvet. 128, Oppiis Helleviis ; CIL. i, 146, Opi. Sanfius.

Epius. CIL. i, 1249, Ep. Popidius, Pompeii ; 193, Epius Fulvius.

Mamercus was an Oscan praenomen according to the statement of
Paul, ex Fest. p. 131 : Mamercus praenomen Oscum est ab eo quod
hi Martem Mamertem dicunt. It passed into Latin as a cognomen ;
cf. CIL. ii, 1475, et al. The only case where it seems to be a prae-
nomen is Fast. Cons. Cap. a. U. c. 349, M' Aimilius Mam. f. M. n.

The Origin of Roman Praenomina. 141

Mamercinus. In Livy, iv, 16, the name appears as Mam. Aemilius,
but in Diod. xi, 65, as Acvkios AZ/luAxos Ma/atp/cos.

Besides these there is a considerable number of other Oscan prae-
nomina which no one has even suspected of being Roman, and which
never occur in Latin inscriptions. These may nearly all be found
in the index to Zvetaieff, Inscript. Ital. Infer.

The Umbro-Sabellian states of the north shore of Italy contributed
a few to our list of sixty-four. The most certain examples are these :

Salvius. This praenomen is clearly non-Latin. I have preferred
to class it as Umbrian rather than Oscan for several reasons : the
majority of inscriptions containing it are found in Umbrian territory;
it is not found as a praenomen in the Oscan inscriptions, and the
corresponding gentile, Salvius, occurs comparatively rarely in Oscan
territory (CIL. x, thirteen times); forms of the adjective salvus, from
which it is probably derived, are very common in the Umbrian tablets.

CIL. i, 1 41 4. Sal. Egnatius. Umbria.

" 1420. Sal. Camurius. Picenum.

" 1280. Sa. Septimius. Paeligni.

" 1286. Sal. Cominius. "

" 183. Sa. Magius. Marsi.

" 184. Sa. Flavius. "

" 1 141. Sal. Statius. Praeneste.

Plautus. ClZ.i, 116, Pla. Magolnius, Latium ; 191, PI. Specius
(unknown.)

The statement of Paulus is of value in locating Plautus: Fest.
p. 238, Umbri pedibus planis . . ; Paul, ex Fest. p. 239, Ploti appel-
lantur qui sunt planis pedibus. Unde et poeta Accius, quia Umber
Sarsinas erat, a pedum planitie initio Plotus postea Plautus est dic-
tus. The poet's name, as Ritschl has finally restored it, appears as
T. Maccius Plautus ; but when we consider that the Italian peoples
outside of Rome did not, according to the evidence of our inscrip-
tions, regularly possess cognomina, and that Plautus, or what appears
to be such, does actually occur as an Italian but not as a Roman
praenomen, it is not too rash to assume that the poet's original name
was Plautus Maccius, and that upon his coming to Rome he changed

142 George Davis Chase.

the unfamiliar praenomen Plautus to a cognomen. For a very simi-
lar instance, see Statins, p. 137; only in the case of Caecilius we are
not told that he assumed a praenomen, as Plautus would seem to
have done. We do not, however, know of anything to prevent a
foreigner from taking an assumed name at Rome if he chose. In
selecting Titus as a praenomen Plautus chose a name that was
familiar both in his native country and at Rome.

Pupius. CIL. i, 1423. Pup. Clodius. Picenum.

Nero. CIL. i, 1412, Ner. Capidas ; 1415, Ner. Egnatius ; 1417,
Ner. Poinisius; Buecheler, Umbrica, p. 172, Ner. T. Babr. ; Vois.
Ner. Propartie, all from Umbria. Suet. Claud. 1, Patrem Claudi
Caesaris, Drusum, olim Decimum mox Neronem praenomine ....
Suet. Tib. 1, Gens Claudia orta est ex Regillis, oppido Sabinorum
.... inter cognomina autem et Neronis assumpsit, quo significatur
lingua Sabina fortis ac strenuus. Gell. xiii, 23, Id autem, sive
' Nerio ' sive ' Nerienes ' est, Sabinum verbum est, eoque significatur
virtus et fortitudo. Itaque ex Claudiis, quos a Sabinis oriundos
accepimus, quis erat egregia atque praestanti fortitudine ' Nero '
appellatus est. We are not surprised to find that the Sabines and
Umbrians used the praenomen Nero in common, as their territories
were adjacent and their dialects closely allied. The fact that the
name when brought to Rome ceased to be employed as a praenomen,
but was adopted among the cognomina, lends support to our Plautus
theory and may throw light on a similar transference of other names;
cf. the case of Statius Caecilius cited above (Gell. iv, 20, 12). Nero
is certainly connected with the Umbrian nerf i nobles'; Skt. nara or
nr ; Gk. dvrjp.

Sertor. CIL. i, 1097, Vib. Vedius Sert. f . ; 1412, T. Mimesius
Sert. f. Incert. Auct. 4 counts Sertor in his collection of prae-
nomina, and explains it thus: Sertor qui per sationem natus erat
appellatus est, — an extremely unsatisfactory sentence for those to
reflect upon who are inclined to accept the position taken by the
Incert. Auct., that praenomina arose from circumstances of birth.
Fest. p. 340, Sertorem quidam putant dictum a prendendo, quia cum
cuipiam adserat manum, educendi eius gratia ex servitute in liberta-
tem, vocetur adsertor ; cum verisimilius sit dictum qui sereret quid,
ac potius adsertorem a serendo cepisse nomen, cum aliquem serat

The Origin of Roman Praenomina. 143

petendo in libertatem eandem qua ipse sit, id est iungat, quia fruges
cum seruntur, terrae iungit. Quod totum Verrius am6dvu>s introduxit.
Sertor once established as an Umbrian praenomen directs us to that
dialect for a derivation, and a very satisfactory one is at once sug-
gested by the verb * serlum (=servare). Sertor, then, standing for
*seritor by a formation similar to that of habitum from habere, would
mean servator ' the preserver.' Compare with this such names from
other dialects z&Numitor 'the arranger,' Tutor Cloilius, Fertor Resius
(Incert. Auct. 1, Kempf, 1854, ab Aequiculis Sertorem Resium qui
ius fetiale constituit ; in his new edition, however, Kempf follows
the MS. and reads Fertorem). We notice at once that such names as
Salvius, Nero, Sertor, etc., are quite distinct in character of meaning
from our classes of Roman cognomina. In establishing those
classes we were obliged to point to Nero as a possible exception to
the system. By assigning to it an Umbrian origin we find at once
confirmation of our suspicion and support for our classification of
cognomina.

Pet., Petr. CIL. i, 1287, from Paeligni, reads L. Ofdius L. f. Pet. n.
We have no means of interpreting the praenomen Pet. In a similar
case, Zvet. 3, Petr., Deecke has guessed Petrus.

A small number of names, unquestionably cognomina, occur
rarely and in late inscriptions in the position of praenomen. Among
these may be mentioned Caesar (Incert. Auct. 3, quae olim prae-
nomina fuerunt nunc cognomina sunt, ut . . . . Caesar ; cf. Kempf,
1854, p. 744, note), Cossus, Taurus, Cordus, Geminus, Denter. Gel-
lius also is found in the position of praenomen in CIL. iii, 606 and
871, but no one, I think, would urge that it represented more than
the imperial confusion of names.

The Roman historians and early records supply us with about a
dozen names used as praenomina which do not appear in use in his-
torical times. These are partly from the legendary history of Rome
and often appear in the name of only a single individual. From the
time of Varro down they have occasioned much speculation and dis-
cussion among antiquarians and etymologists, — perhaps more than
their importance warrants. They are as follows :

lulus, the fabled son of Aeneas, who was the legendary founder of
the gens lulia. Livy (i, 3) hints that his name was lulus Ascanius.

144 George Davis Chase.

lulus appears as a cognomen in Diod. xi, 65, Acvklos 'IovAios "IovXos.
The name is very obscure, but may be, as Lindsay suggests, derived
from lovis. A * lovilus became * Ioilus, which in turn changed to
lulus; cf. the Oscan iovilo. Diminutives in -/us are common as
cognomina and are also frequent in our provisional list of praenom-
ina ; as Aulus < * Avolus, Proculus, Tullus, Paulus, cf . Gk. iraTs <
*ira/:-iS-s. We can hardly imagine lovis entering into a personal name
except as one member of a compound. A * lovi-carus, or similar
compound, became simplified to a single member, which then took
the diminutive ending -/us; as U/fi/as, ©paVvAAos, Celt. ToutU/us,
Slav. Dobri/o.

Ancus, found only in the name Ancus Martius. Paul, ex Fest. 19,
Ancus appellator qui aduncum bracchium habet et exporrigi non
potest. Incert. Auct. 4, Ancum praenomen Varro e Sabinis trans-
latum putat. Valerius Antias scribit quod cubitum vitiosum
habuerit qui Graeci vocatur ayntov. Ancus appears to be the noun
from which the diminutives ancu/us 'servant,' ancilla 'maid' are de-
rived. Ancus Martius has therefore been interpreted as the ' servant
of Mars.' The derived gentile Ancilius is found in CIL. i, 1144.

Numa, found only in the names of Nutna Pompilius and JVuma
Marcius (Liv. i, 20, 5). In the Greek historians Numa appears as
No/u,as; cf. Dion. Hal. ii, 59-76. The gentile Pompilius joints to a
non-Latin origin, and the family undoubtedly belongs to the many
which migrated in the earliest times to Rome and left their traces in
their family names and even in words in the Latin language which
can only be explained as borrowings from a non-Latin dialect, as
bos, popina, etc. Tradition also, according to which Numa Pompilius
came from the Sabines, supports this view. Serv. ad Aen. vi, 808,
orta est bona Pompilii fama quod esset apud Cures, civitatem
Sabinorum. Unde etiam Numa dictus est cb-o ra>v v6/xo>v. In gen-
eral, the statements of the ancients with regard to the origin of fam-
ilies are fairly reliable, their etymologies of names too often in
evidence against them. Still it seems most likely that Numa,
together with Numitor and Numerius, is derived from the root which
appears in vo/tos, vi^ta, numerus, etc., meaning ' arrange,' ' order.'

Hostus. Liv. i, 12, Hostus Hostilius fought on the side of the
Romans against the Sabines ; he was grandfather of the king, Tullus

The Origin of Roman Praenomina. 145

Hostilius, iv, 30. Hostus Lucretius Tricipitinus was consul in 327
a. U. c, as also appears from the Fasti Consulares. Incert. Auct. 4,
Hostus praenomen fuit in eo qui peregre apud hospitem natus erat,
idque habuit Lucretius Tricipitinus, collega L. Sergii. It is hard to
believe that Hostus stands directly for kostis, which was and remained
an -i stem ; cf. Goth, gasts, OS1. gosti. That it may represent the
abbreviated form of some old Gompound of hostis meaning ' stranger '
(Cic. de Off. i, 12, 37) is highly probable. Cf. Gast-, a very common
member of OHG. and OS1. compound personal names, as in OHG.
Hadu-gast, Koni-gast, Lindi-gast, Gast-hart; in abbreviated form,
Gasto, Gastilo; OS1. Gosttrad, dim. Gostilo. (Cf. Fick, die Griech.
Personennamen, 1874, p. lxvi.) In historical times Hostus survived
in the gentiles Hostius and Hostilius.

Proculus is found as a praenomen in Liv. i, 16, Proculus Iulius ;
ii, 41, Proculus Verginius ; iv, 12, Proculus Geganius Macerinus,
consuls; in the Fast. Cons, for the years 268, 319, 314; also Diod.
Sic. xi, 1, and xii, 49, HpoKkov Qvcpymov TpiKocrrov; xii, 36, IIpoK\ov
T€ydvtov. Incert. Auct 3, quae olim praenomina fuerunt, nunc cog-
nomina sunt, ut . . . . Proculus. That it was later used as a cogno-
men appears from Tac. Hist, i, 87, Licinius Proculus. Paul ex Fest.
p. 225, Proculum inter cognomina eum dicunt qui natus est patre
peregrinante a patre procul. Proculos sunt qui credant ideo dictos
quia patribus senibus quasi procul progressis aetate nati sunt. Pro-
culus is nothing more than the diminutive of procus, a suitor, or per-
haps with an older meaning of the word, for we learn from Fest. p.
249, that anciently proci was equivalent to proceres. It appears,
therefore, to belong to the class of words used as cognomina. It is
the basis of the gentile Procilius.

Tullus appears as a praenomen as early as Tullus Hostilius.
Elsewhere in Livy it is used as a cognomen; cf. iv, 17, Cloelius
Tullus (Plin. N.H xxxiv, 6, Tulli Cloeli) ; ii, 35, Attius Tullus; 37
and 38, Tullus. As a praenomen it is also found in Cic. Phil, ix, 2,
Tullus Cluvius ; CIL. i, n 20, Tul. Tullius Tul. f. (Tibur) ; also as
the basis of a gentile as early as Servius Tullius, and in historical
times M. Tullius Cicero, etc. Incert. Auct. 4, Tullus praenominatus
est ominis gratia quasi tollendus, o littera in u conversa. The word
seems to show the diminutive suffix -lo. Whether the first part

146 George Davis Chase.

stands for tul (cf. tuli, tetuli, etc.), as is generally received (Vanicek,
p. 296) or whether the first / comes by assimilation from some
other consonant it is difficult to say. In the latter case we might
think of Tudulus (cf. Tudicius, Tuccius) > *Ttidlus > Tul/us, or
perhaps * Tutu/us (Osc. tovto ' people ') > * Tutlus > Tuttus, an
abbreviated compound name (cf. Celt. Toutillus, etc.). Such assimi-
lations are hard for classical Latin, but might have taken place more
easily in the earlier or plebeian speech. On the whole, Tullus
' little supporter ' is the safest etymology. Whether that, however,
is to be considered as an original cognomen or the shortened
form of some compound, as opitulus 'help bringer,' is again a
question which we are not able to decide.

Vopiscus. Liv. ii, 54, 3, Vopiscus Iulius ; Fast. Cons. a. U. c. 281,
346, 353, 361. Incert. Auct. 4, Vopiscus, qui in utero matris gemi-
nus conceptus, altero abortu iecto, incolumis editus erat. Plin. N. H.
vii, 10, Vopiscum appellabant e geminis qui retenti utero nascerentur
altero interempto abortu. Namque maxima etsi rara circa hoc mira-
cula existunt. Non. 557, Vopiscus qui ex duobus conceptis uno
abortu excluso ad partum legitumum deducitur. Sol. i, 69, E gemi-
nis si remanente altero, alter abortivo fluxu exciderit, alter qui legi-
tume natus est Vopiscus nominatur ; cf. Isid. Orig. 9, 5. The
agreement of these various late writers means, I think, nothing more
than a common source of information. It does not seem physiologi-
cally probable that such births occurred frequently enough to give
rise to a special personal name to describe them. It is hardly pos-
sible, for example, that several consuls should have been so born or
that the name could become extended in use so long as its meaning
was borne in mind. If, however, the meaning of the nanie was, as
early as 350 a. U. c, already forgotten, we wonder how it survived
till the time of Pliny. We are inclined from the, length of the word
to consider it a cognomen in origin occasionally used as praenomen
(cf. chap, vii) rather than the abbreviated form of a compound
name which would rarely contain three syllables.

Agrippa appears as a praenomen in the name Agrippa Menenius
(Liv. ii, 16; iv, 11, 13; Diod. xiii, 7), in Agrippa Menenius Lana-
tus (Liv. iv, 44, 47), in Agrippa Furius (Liv. v, 32 ; ii, 66: Diod.
xii, 30), and is also found in the Fast. Cons, of a. U. c. 251, 312,

The Origin of Roman Praenomina. 147

302, 315, 335, 337, 308, 363. The earliest record of the name is in
the case of Agrippa Silvius, one of the early Alban kings (Incert.
Auct. 1), who appears in Livy (i, 3) as Agrippa simply. That
Agrippa was later used as a cognomen is shown by the name M.
Vipsanius Agrippa, of the time of Augustus. It also appears as a
cognomen in the gentes Asinia, Fonteia, Hateria, lulia, and Vibulana.
Incert. Auct. 3 also classes it among the, names ' quae olim praeno-
mina fuerunt, nunc cognomina sunt.' On the meaning of the word,
Non. 557, says, ' Agrippae qui cum labore matris eduntur, hoc est
per pedes contra naturam, non per caput, quasi ab aegro partu.' A
wilder etymology would be hard to seek. This explanation is evi-
dently copied from Plin. N. H. vii, 6, 1, in pedes procidere nascen-
tem contra naturam est, quo argumento eos appellavere Agrippas ut
aegre partos ; qualiter et M. Agrippam ferunt genitum, unico prope
f elicitatis exemplo in omnibus ad hunc modum genitis ; cf . also Gell.
xvi, 16 ; Sol. i, 65 ; Serv. ad Aen. viii, 682. Agrippa is probably a
Greek name, compounded of dypds and tmros. It does not occur in
Greek sources, but Fick2, Griech. Personennamen, p. 45, cites the
similar names 'Aypo-Aecov, 'Ayp-oirai ; p. 7, 'Aypd-A.as and Ae-
aypos. The form in Greek would be'Ayp-fn-n-os and it is necessary
to suppose that the Italians who borrowed the name changed it to
an -a stem, on the analogy of some of their own names, as Osc.
Maras, Tanas, Lat. Catilina, Seneca, etc. It is likely that the name
found its way to Rome from some of the Greek colonies of Magna
Graecia or Sicily through some early emigrant, and, although at first
used indifferently as praenomen or cognomen, finally settled into its
proper place as cognomen.

Volero, Volesus, Volusus, and Valesus are apparently dialectic
variations of the same name. Volero occurs, Liv. v, 13, ii, 55, in
Voleronem Publilium. Later the same man is repeatedly spoken of
as Volero alone, as if the name was considered a cognomen rather
than a praenomen. In Diod. xiv, 54, he appears as OiaXipios HoirXC-
Aiog; cf. also Fast. Cons. a. U. c. 354, 355. Volesus appears in Liv.
i, 58, P. Valerius Volesi f.; in Dion. Hal. ii, 46, OidAeo-og (Jacoby,
OidAoo-cros) OiaXeptos ; and in Fast. Cons. a. U. c. 294, 298, 339,
350. This Volesus, we are told by Dion. Hal. I.e., was a powerful
Sabine who came to Rome with Titus Tatius (see also Incert. Auct.

148 George Davis Chase.

1) and became the founder of the gens Valeria; cf. Ovid, ex Pont.
iii, 2, 105, quos Volesus patrii cognoscat nominis auctor; Sil. Ital.
ii, 8, Poplicola ingentis Volusi Spartana propago; Juv. Sat. viii, 182,
et quae turpia cerdoni, Volesos Brutumque decebunt. CIL. i, 187,
Va. Condetius is probably to be rendered Valesus. The form Valesus
is also preserved in Valerius, beside which we have the gens Volusia.
As to the origin of the name, the Incert. Auct. 4 says, 'Volero in
praenomen abiit quod volentibus nasci liberi parentibus videbantur.'
It is most likely that the name is derived from the root of volo, and
may well be part of a compound name. We may suppose a neuter
-es stem noun, *vefces 'desire,' developing an a or 0 vowel by the
influence of the /; cf. Lindsay, Lat. Lang. § 92. The Latin divini-
ties Vol-u-mnus, -a, 'the well wishing one,' furnish a good parallel
for the use of sj vol. Compare, for the compound, such names as
Greek Eu-/?ou\os; Germanic, Wili-frid, Wili-muot, Wilt-rat, and
many others.

Opiter. Liv. ii, 17 and 54, Opiter Verginius; Fast. Cons. Cap.
a. U. c. 352, [L. Vi]rginius L. f. Opetr. n. Tricost. Esquilin. CIL. i,
146, On • SAVFIO • I • I, from old Latium, has been generally inter-
preted as Opiter, although I am inclined to take it rather as the
Oscan praenomen Opius; cf. Zvet. 128, Oppiis Helleviis, and
examples of Opius as an Oscan gentile, Zvet. p. 136. Dion. Hal. v,
49, 'Omrtop Ovepytvios ; Diod. xii, 73, 'Owirepos AovKpTJTws ; Zonar.
vii, 13, 'OiriT<opios Ovepymos. A derived gentile appears in Liv.
xxxix, 1 7, in the Faliscan name L. Opiternium. For the meaning
of the word we have some interesting etymologies from classical
sources, as follows : Paul, ex Fest. p. 184, cuius pater avo vivo mor-
tuus est, ducto vocabulo aut quod obitu patris genitus sit, aut quod
avum ob patrem habeat, id est pro patre; Incert. Auct. 3, Opiter
vocabatur qui patre mortuo, avo vivo gignebatur; Quint, i, 4, 25,
et ex casu nascentium ; hie Agrippa et Opiter et Cordus et Postumus
erunt. Following these authorities, modern scholars are accustomed
to explain the word as a compound of avus and pater, and to cite as
parallels Diespiter, luppiter, Marspiter. So Stolz and Schmalz, in
I. v. Muller's Handbuch*, ii, p. 272 : 'Im Schriftlatein haben wir
o(u)=au in opiter = *av(i)-piter'; Zimmermann, Neue Jahrbticher,
1896, pp. 419-20 : 'Opiter das ich fur einen vocativ und zwar aus

The Origin of Roman Praenomina. 149

av(e) pater entstanden ansehe (avpiter, aupiter, dpiter). das kind
wollte, wie bei uns den grosspapa bzw. o-papa vom papa, so den
avus pater vom pater unterscheiden. die vocativform, weil vom
kinde bzw. dem kinde gegenuber haiifiger angewandt, wurde die
herschende, wie in Ju-piter, Juppiter, vgl. Z«i) iraTcp, Brugm. Grund-
riss, i, s. 464. der iibergang von Aupiter zu Opiter lasst sich eben
aus der f amiliaren sprache leicht erklaren ; vgl. Olus neben Aulus.'
Other recent views are presented in Stolz, Hist. Lat. Gram, i, p. 211,
none of which meet with the present writer's approval. We have
first to notice that *Avipiter ' possessing a grandfather as father ' is
not to be classed as a compound with luppiter 'father Zeus,' and
that it would be a difficult compound in Latin, which restricted itself
to compounds of the simplest and most transparent signification.
Secondly, by all analogies 0 arising from au is long; but uniformly
Opiter is written with o in Greek and appears with short 0 in Sil.
Ital. x, 33, sternuntur leto atque Opiter quos Setia colle. These
considerations seem to me fatal for the explanation of Opiter as
meaning ' one who has a grandfather in the stead of a father.' Nor
do I believe that pater appears in the word; rather is it to be
derived from the root op 'help,' as it appears in the names of the
divinities Ops and Copia (= co-op-ia), and in the gentiles Op-lmius
and Op-isius. The -i- is a developed connecting vowel. For the
•t-, cf. op-tare, op-tumus; for older opitumus (cf. CIL. i, 1016,
opitumd). A feminine name of similar formation occurs, CIL. viii,
476, in Caecilia Opita. The -er is more obscure; it may be com-
pared with the ending in the cognomen Dent-er, and in the noun
accipiter < *acu-pet-er ' the swift-flyer,' — a comparison which is
favored by the similarity of their declension ; cf. Priscian, 6, p. 229,
Keil, inveniuntur tamen apud vetustissimas haec ancipitis genitivi :
"hie accipiter, huius accipiteris," et " accipitris," "Opiter Opiteris"
et " Opitris." Opiter, therefore, means ' the helper,' and is very
similar in kind to such names as Sertor 'the preserver,' Numitor
' the arranger,' Fertor ' the supporter,' etc.

Kaeso lingered longer in use as a praenomen and was the only
one of the early class that had a generally recognized abbreviation,
namely K. Examples of its use in the gens Quinctia are common
in Livy and the Fasti. In inscriptions it appears in K. Fabricius,

150 George Davis Chase.

CIL. i, 107, an early inscription of Latium. On the meaning of the
word Paul, ex Fest. p. 57 says : Caesones appellabantur ex utero matris
exsecti. Plin. N. H. vii, 9, . . . sicut Scipio Africanus prior natus
primusque Caesarum a caeso matris utero dictus, qua de causa et
Caesones appellati ; cf. Isid. Orig. ix, 3, 12. Marquardt, Rom.
Alterthumer, iii, p. 441, and Mommsen, Rom. Forsch. p. 17, under-
stand Kaeso of the lashing (caedere) which the Luperci did. But it
seems preferable to take Kaeso as an -0, -onis adjective connected
with caesius, and referring to the color of the eyes or complexion; cf.
Vanicek, Etym. Worterbuch, p. 1002, especially the names Caesonius,
Caesulla (= * Caeson-la), and Paul, ex Fest. p. 136, caesullae a
caesiis oculis. Kaeso will then be in origin a cognomen pure and
simple.

With Kaeso our list of these obsolete praenomina is concluded.
We have found that these, if they survived till historical times, were
used only as cognomina, and that in origin they belonged mostly to
those classes of names into which the Roman cognomina fall. Only
three — Hostus, lulus, and Volero — can with any degree of proba-
bility be looked upon as survivals of original Indo-European names.
And yet we will admit, for the evidence points in that direction, that,
in the period for which they are recorded, all these names were alike
considered and employed as genuine praenomina.

We have next to consider a class of names which certainly arose
as cognomina and were so used in classical times, but also occur
with more or less frequency as praenomina.

Faustus is found oftenest as a cognomen, but occasionally as a
praenomen ; cf. Val. Max., Kempf. 1854, p. 745. The Incert. Auct.
derives Faustus as a praenomen from favor.

Faullus occurs CIL. i, 473, Paullus Lepidus ; 799, Paul. Fabius;
and occasionally in literature.

Fostumus. CIL. i, 79, Post. Antestius ; 1089, Postumus Sulpicius ;
1412, Post. Mimesius ; and a number of times in literature. The
Incert. Auct. classes it among the praenomina that became cogno-
mina.

More important are the numerals. These all, from Frimus to Deci-
mus, are used as cognomina. The first four — Frimus, Secundus,
Tertius, Quartus — never appear as praenomina until after republican

The Origin of Roman Praenomina. 151

times. Septimus appears as praenomen, Liv. xxv, 37, L. Marcius
Septimi Alius, and xxviii, 28, 13, Septimum Marcium, but xxxii, 2, 5,
L. Marcio Septimo. In other authors he is merely L. Marcius.
The gentiles, also, from the numerals, as Sextius, Septimius, Octavius,
Nonius, Quinctius, etc., are of very frequent occurrence. Only three
of the numerals were regularly and frequently used as praenomina
at all periods, namely, Quintus, Sextus, and Decitnus, abbreviated as
Q., Sex., and D. In CIL. i, occur as such Quintus, 241 times,
Sextus, 57 times, Decitnus, 22 times. But although we find so
marked differences in the use of the numerals as names, it is impos-
sible to believe that this use did not arise in the same way for all ;
that is, that they were at first used to distinguish children by the
order of birth. But when we find them as praenomina in historical
times it is evident that they no longer referred to order of birth. A
very clear proof of this is seen from the gens Iunia (Drumann, iv, 1),
which was represented in the consulship in three successive genera-
tions, B. C. 325, 292, and 264 by a D. Junius Brutus, and again by
two consuls of the same name in 138, and 77, the latter of whom
had a son D. Junius Brutus. In all, there were eight consuls
named D. Junius Brutus. That these were all tenth sons is out of
the question.

The praenomina which remain to be considered are those which
were commonly used and accepted as such throughout the classical
period of Roman literature and which are almost universally abbre-
viated in Roman inscriptions. Of these, if we may believe a very
probable tradition, a small number came to Rome with early immi-
grants from other parts of Italy, but at a period so remote that in his-
torical times they were regarded as genuine Roman names and had
already acquired a considerable diffusion among the Roman people.

Appius, the most restricted of these, is found as a praenomen in
the Claudian gens as early as a. U. c. 259, when Appius Claudius
Sabinus was consul, and later in other gentes, as Annia, Modia, Popi-
dia, and Iunia. The name Appius is reported to have been brought
to Rome from the country of the Sabines. Liv. ii, 16, namque
Attus Clausus, cui postea Ap. Claudio fuit Romae nomen . . . . ab
Regillo magna clientium comitatus manu Romam transfugit. iv, 3,
Claudiam certe gentem post reges exactos ex Sabinis non in civitatem

152 George Davis Chase.

modo accepimus, sed etiam in patriciorum numerum ? Suet. Tib. 1,
Patricia gens Claudia orta est ex Regillis, oppido Sabinorum. Inde
Romam recens conditam cum magna clientium manu commigravit,
auctore Tito Tatio, consorte Romuli, vel, quod magis constat, Atta
Claudio, gentis principe. Serv. ad Aen. vii, 706, Clausus, Sabinorum
rex, post exactos reges, ut quidam dicunt, cum quinque milibus
clientum et amicorum Romam venit, et susceptus habitandam partem
urbis accepit : ex quo Claudia et tribus est et familia nominata. Cf.
also Sil. Ital. xv, 546 ; Tac. Ann. iv, 9 ; Plut. Popl. 21 ; Coriol. 11 ;
Dion. Hal. v, 40. The use of Appius in the Claudian gens was
peculiar, in that it seemed to partake of the character of both prae- •
nomen and gentile. It appears with the force of a gentile in such
expressions as via Appia, aqua Appia, forum Appii, etc., and in the
case of several pairs of brothers named Appius Claudius. For a
fuller treatment of this subject see Mommsen, Horn. Forsch. p. 25.
A further discussion of the origin of the name Appius would take us
off of Latin soil, and into a consideration of non- Roman names to an
extent to which we do not at present care to go. An interesting
article by Zimmermann on the meaning of Appius may be found in
the Neue Jahrbucher, 1896, p. 420.

Titus, in much the same way, but attaining a somewhat wider ex-
tension of use at Rome, came from the Sabines. . Incert. Auct. 6,
Titus a Sabino nomine Tito * fluxit, Appius ab Atto, eiusdem regionis
praenomine. Titus Tatius, according to all traditions, was the king
of the Sabines who first made terms with the Romans (Liv. i, 10).
Whatever view we may hold of the story, we must attach weight to
that part of the tradition which declares Titus to be a Sabine name.
This view is further borne out by the fact that Titus occurs very fre-
quently as a praenomen in the Latin inscriptions from Umbria and
the Paeligni, and also in the Umbrian inscriptions (Buecheler,
Umbrica, p. 172) in names which are certainly native to that region.
Titus is regularly abbreviated by T., except CIL. i, 1292, where we
find Tit. Of all names of foreign origin Titus gained the widest cir-
culation at Rome, a fact undoubtedly due to its early introduction
and the distinction of those who first bore it. Cf. Zimmermann,
Neue Jahrbucher, 1896, p. 420, for a discussion of the meaning of
the word.

The Origin of Roman Praenomina. 153

Ten of our provisional list of sixty-four praenomina yet remain.
Four of these — Gnaeus, Aulus, Tiberius, and Spurius — have the
appearance of having first arisen as early cognomina.

Gnaeus (Gnaivos') is an older form of the common noun naevus
' birthmark.' So Paulus ex Fest, p. 96, Gneus et corporis insigne et
praenomen a generando dicta esse, et ea ipsa ex Graeco ytyvcaOai
apparet. On the various spellings of the word, cf. Incert. Auct. 5,
Gnaevus ab insigne naevi appellatus est : quod unum praenomen
varia scriptura notatur : alii enim Naevum [cf. gens Naevia], alii
Gnaeum, alii Cnaeum scribunt. Qui G littera in hoc praenomine
utuntur antiquitatem sequi videntur, quae multum ea littera usa
est. Olim enim dicebatur frugmentum, nunc frumentum, et forgtis,
non fortis, et gnatura, non natura : ergo etiam qui in corporibus gigni
solet gnaeus appellabatur.

Spurius is sufficiently explained as the Latin adjective (=v66o$,
cf. Ndtfwv, Herod, vi, 100), in the sense of 'born out of legal wed-
lock.' Fest. p. 174, nothum Graeci natum ex uxore non legitima
vocant, qui apud nos spurio patre natus dicitur, quod Ser. Tullius
qui Romae regnavit natus est ex concubina Spurius *Tulli tributis,
etc. Such an explanation, so near to seek, and so much in keeping
with the Roman custom of naming, is much to be preferred to a more
distant one, as, for example, Deecke's attempt to derive Spurius from
the Etruscan spure (Deecke, Etrusk. Forsch. u. Studien, ii, p. 43).

Aulus. The explanation of the Incert. Auct., ' Auli qui diis alen-
tibus nati sunt,' cannot possibly be correct, as alo is built on the
simple root al, and could not by any law produce an aul-. Neither
are we in favor of deriving it from the Etruscan avile, with Deecke,
Etrusk. Forsch. u. Studien, v, p. 139. Two choices remain, — to derive
the name from olla (cf. Paul, ex Fest. 20, aulas antiqui dicebant quas
nos dicimus ollas, quianullam litteram geminabant), making it in some
way to refer to the personal appearance (cf. English ' Tubby ' ), in
which case we should expect the name to be Aula ; or preferably to
explain it as standing for avulus ' little grandfather ' (cf. English
'uncle,' French 'pere'). Since I reached this conclusion, there
has appeared an article by Zimmermann, Neue Jahrbiicher, 1896, p.
419, presenting the same view. We find a derivative of Aulus in the
gentile Avilius. Avitus is used as a cognomen.

154 George Davis Chase.

Tiberius can only be considered in connection with the related
place-names, Tiberis, the river, Tifernum, a town of Umbria, and
Tifernus, a mountain and stream in Samnium. Perhaps also the
town Tlbur is related. The cognomen Tibullus seems to stand for
*Tiber-lus. Paul, ex Fest, p. 366, derives the river from a personal
name : Tiberis fluvius dictus a Tiberino, rege Albanorum, quod in
eo cecidisset. Tibris a Tibri, rege Tuscorum. This is not so
probable as that the personal name was derived from that of the
place. So the Incert. Auct. 6, Tiberii vocitari coeperunt qui ad
Tiberim nascebantur. More probably the meaning was 'one who
lives near the Tiber.' That personal names were early derived
from place-names is shown by the cognomina in the first book of
Livy which refer to localities, as Coriolanus, Capitolinus, etc.

The six praenomina that remain to be considered — Servius,
Manius, Publius, Lucius, Gaius, and Marcus — differ radically from
those of the cognomen class ; they are not simple words out of the
language, but are formed with derivative suffixes, — the first five
with -to-, Marcus with -co-. Again, we shall see, when we look more
closely at their meanings, that, they are formed from words which do
not criticise individual peculiarities, which are never derogatory or
familiar, as is the case with the cognomina. On the contrary, they
express either praise or a wish or hope for moral or physical dis-
tinction that might be intended as an omen for future greatness in
the career of the man. In short, they are exactly the kind of names
that, as was pointed out in a previous chapter, were employed in
the original Indo-European system ; only in every case they appear
in the abbreviated form, showing only one member of an earlier
compound, with some derivative ending attached. Owing to their
extreme antiquity, their origin was entirely obscured in historical
times, and they were no longer distinguished from the other names
in use. Hence we are not surprised to find the Latin antiquarians
widely erring as to their derivation and meaning.

Servius is to be derived from a root ser ' bind ' ; hence, as seems
to be its force in names, 'to collect carefully,' 'to preserve.' The same
root is seen in the Umbrian name Ser-tor and in the verb seritu ' pre-
serve.' In Latin the root, when used in the sense of 'preserve,'
seems always to have taken a -uo suffix. Thus we have ser-vo-s,

The Origin of Roman Praenomina. 155

which means doubtless the keeper, preserver of the owner's goods.
From servus comes the denominative ser-va-re, to preserve, and
later, from servus in its classical sense, ser-vl-re, to serve. Servius,
then, stands for a compound having servus as one member, for a

Servo- -\- , or + servus; a 'preserver of heroes (?) ' or

'country,' etc., or 'God (?) protected,' as in Greek 'AAc'l-avSpos.
After servus had come to acquire the odious connotation of ' slave,'
it is easy to see how the praenomen Servius, which was plainly in
some way connected with it, fell into disuse, and we are not surprised
to find it comparatively rare in classical times. Having thus dis-
posed of Servius, it is only of passing interest to note the opinions
of the ancients. Incert. Auct. 6, Servius quod mortua matre in
utero servatus est. Dion. Hal. iv, 1, expresses a different view :
to o« koivov Kal irpoo-rjyoptKov, 'Stpoviov, im tiJs iSt'as rv^rp, on SouAevovo-a
erCKev avrov. at/ 8 av 6 Sepovtos, as ttjv 'EAAjjnK^v SioXcktov ;u,£Ta/?i/t?a£o-
fievos, AovAxos. And this view seems to be hinted at by Fest. p. 174,
in speaking (like Dionysius, I.e.) of Servius Tullius in the very
corrupt lines, Ocilisiam corniculam captivam eum susceptum matre
servientem.

Manius. Varro, L. L. ix, 60, forsitan ab eo qui mane natus diceretur,
ut is Manius esset. Paul, ex Fest. p. 1 48, Manius praenomen dictum
est ab eo quod mane quis initio natus sit. Incert. Auct. 5, Manii qui
mane editi erant, vel ominis causa quasi boni. Manum enim an-
tiqui bonum dicebant. The existence of the old adjective manus,
meaning 'good,' is well attested. So in Fest. p. 146, manuos in
carminibus Saliaribus Aelius Stilo significare ait bonos : et Inferi di
Manes pro boni dicuntur a suppliciter eos venerantibus, propter
metum mortis, ut immanes quoque pro valde .... dicuntur. Serv.
Aen. i, 139, Immania aspera : manum enim antiqui bonum dicebant,
sicut supra dictum est, unde et mane dicitur ; quid enim melius ?
et per antiphrasin ' manes ' Inferi, quia non sint boni. The word
was also preserved in the names of Latin divinities, as Cerus manus
' the good Creator,' Mana the goddess of death and birth, Maniae
(cf. Fest. p. 145 ; Macrob. Sat. i, 7, 34), and Summanus, the old Latin
divinity of nocturnal lightning, Varro, L. L. v, 74. It is much the
most likely that Manius represents a compound of this old adjective
manus, and not a formation from manus in the unattested derived

156 George Davis Chase.

meaning of 'early in the morning.' We have only to point to the
fact that various words meaning ' good ' are among the very com-
monest in the original compound names, and our position seems
quite unassailable. Compare such names as 'Ayaflo-xAiJs, 'Aya0o-
Aas, ' Ay a.6 6-8 wp os, etc. But Manius was not very extensively in
use as a praenomen. Perhaps this was due to the fact that the
name carried with it an unpleasant, uncanny sensation because of
its seeming connection with Maniae. This idea is hinted at by Fes-
tus in the following passage, p. 145 : Manius Eger .... Nemorensem
Dianae consecravit a quo multi et clari viri orti sunt, et per multos
annos fuerunt ; unde et proverbium : " Multi Mani Ariciae." Sinnius
Capita longe aliter sentit; ait enim turpes et deformes significari,
quia Maniae dicuntur deformes personae, et Ariciae genus panni
fieri ; quod manici appelletur.

Publius. Incert. Auct. 5, Publii qui prius pupilli facti erant quam
praenomina haberent ; alii ominis causa ex pube. Beside populus
there apparently existed in old Latin a noun *poublus built on the
root oipubes, and so closely coinciding in meaning with populus that
it was superseded by it in use. Of the adjectives, however, derived
from these nouns, publicus survived to the exclusion of poplicus (cf .
CIL. i, 196, poplicod). Publius therefore is the contracted form of
a *Publo-carus, or similar name. Cf. I. v. Miiller's Handbuch 2, ii, p.
383 ; Lindsay, Lat. Lang. p. 287. The word populus or * pub/us
(= Greek wkfjOoi) evidently arose on Italic soil as the regular
designation for ' people.' The earlier word, which appears in Oscan
as tovto, Goth, yiuda, Celt, tuath, etc., is very frequently used in
compound personal names, as Celt. Toutio-rix, Germ. Diet-rich (see
also the suggestion for Tullus, p. 146). In Italic the transition to a
new term is explained by the expanded phrase in Umbrian, poplom
totar ^=populum tutae) ; cf. Buecheler, Umbrica, vii, A. 15.

Lucius. The prevailing idea among the Romans was that Lucius
meant one born at daylight. Thus Incert. Auct. 5, Lucii coeperunt
adpellari qui ipso initio lucis orti erant aut, ut quidam arbitrantur, a
Lucumonibus Etruscis (cf. Dion. Hal. iii, 48). Paul, ex Pest. p. 1 19,
Lucius praenomen est eius qui primum fuit, qui oriente luce natus
est ; cf. ibid. p. 148. Varro, L. L. ix, 60, qui mane natus diceretur
ut is Manius esset, qui luci, Lucius ; ibid, vi, 5, Secundum hoc

The Origin of Roman Praenomina. 157

dicitur Crepusculum a crepero, id vocabulum sumpserunt a Sabinis,
unde veniunt Crepusci nominati Amiterno qui eo tempore erant
nati, ut Lucii prima luce. Lucius represents the shortened form
of a compound of the old adjective * loucus. This adjective had
ceased to be used as such in Latin in classical times, but survived
as a noun in loucus ' a cleared grove.' The original meaning was
' bright,' ' shining.' In Greek we find the exact correspondent in
XevKos, and in Lettish in /auks (=* loukos~), which has become spe-
cialized to denote a horse with a white spot in the forehead ; cf. Lat.
calidus. In classical Latin the name was certainly Lucius. There
is, however, some evidence that in the older language it had a differ-
ent form. Thus in the Scipio epitaphs we have, CIL. i, 30,
Cornelius Lucius Scipio, and in 32, Lucfom Scfpidne, where the
requirements of Saturnian verse seem to have obliged the poet to
place the praenomen after the gentile, and to demand a scansion of
Lucius. So Wordsworth, Fragments of Early Latin, p. 398, explains
it, and cites as additional proof the modern Italian Lucia. This,
however, though obscure in origin, can have little weight in the dis-
cussion ; it is impossible to believe that an early * Lucius survived
till modern times against the universal classical Lucius. L. Miiller,
de Re Metrica 2, p. 286, believes that the early form was Luclvus; but
this is in part refuted by the Oscan form of the name, Luvikis or
Luvkis ; cf. Zvet. 128. The meagreness and uncertainty of all this
evidence, added to the fact that any views we may hold of the
Saturnian verse are at least debatable, oblige us to regard Lucius as
the most probable original Latin form. As was common in other
languages, the shortened form took on also other suffixes : thus
beside Luc-io-s we have LUc-ullus, used as a cognomen. In force,
Luco- was probably equivalent to the Germanic berht in such names
as Al-bert, Bert-hold, etc. Etymologically it may be connected with
such names as the Skt. Rukma-bahu, Ruci-deva, Vara-ruci, and the
shortened Ruci, Roci, Rocana, etc.; the Avest. Raocac-caeshman, Vohu-
raocaiin,'Pu>^d.vr] ; Gk. Acvk-iititos, Aevxos, Aev*<ov, Zd-XevKos ;
Celt. Lugoto-rlx, Lug-dunum, etc.

Gaius has certainly lost a medial v, as is shown by the gentile
form Gavius. With that supplied, the etymology is clear. The root
gau 'rejoice' is common in both Latin and Greek, as in gau-deo,

158 George Davis Chase.

gavtsus (=* gau-id-tus*) ya-i'-w, dyav-os, etc. It also appears in the
Greek name Taw-fi.^Srj's, and as a name-forming element is highly
appropriate. The Incert. Auct. 5 saw the correct derivation of the
word, which he explains of course wrongly : Gaii iudicantur dicti a
gaudio parentum. Rather the original compound was meant to ex-
press the wish that the child might come to be the joy of his people,
of Mars, of women, etc.

Marcus, the last and most difficult of the praenomina, does not
mean as the Incert. Auct. 5 explains : Martio mense geniti. The
same root is evidently represented in Marcus and Mars, but the
exact relation of these two words is very hard to determine. Marcus
and also the derivatives Marcius and Marcellus show constantly long
a. Mommsen, Ephem. Epigraph, i, 286, shows that the Greeks
regularly wrote Maapicos, and that they regularly used the double
vowel in no other Latin word. He concludes that the Romans pro-
nounced Ma(h)arcus, which was contracted to Marcus, and compares
Ala, for Ahala ; see also ibid, iv, p. 217. In I. v. Miiller's Handbuch,
ii, p. 281, Maarcus is explained as a case of vowel lengthening
before the combination r + consonant. For Mars there is hardly
any evidence to lead us to believe that the vowel was long. In the
Monument. Ancyr. iv, 21, it occurs with the apex, M ART IS, but in
three other following places (iv, 25 ; v, 42 ; vi, 31) without any sign
of length. The reduplicated forms in Sabine and Oscan, Ma-mer(t)s,
Ma-mer-cus, with vowel weakening in the unaccented syllable, point
to a short vowel. It seems likely that the root appeared in two
ablaut grades in Latin, — as mar in mar-mor, Mar-o Mar-ius,
Mar-(t)s, Ma-mer-(f)s ; and as mar in Mar-cus. It is extremely dif-
ficult to maintain any connection in form between Mars and the
poetical form Mavors, even through the medium of the old Latin
Maurte. Supposing the 0 to have dropped, we have no example of
au becoming a except in a few cases of late Latin ; cf. I. v. Miiller's
Handbuch, ii, 272, where an attempt is made to establish the rela-
tionship, and Corssen2, i, 664. Supposing, too, Mars to stand for
an older Mavors, Mamers would be very difficult of explanation.
The Oscan praenomen Mamercus(ci. Paul ex F. p. 1 3 1 , and for quantity
see Juv. viii, 192) probably stands for Mamertus, being corrupted in
Latin by the analogy of Marcus. We find also the cognomina

The Origin of Roman Praenomina. 159

Mamertus and Mamertullus in use, and in Ovid, lb. 547, Mamerta,
as a masculine. Least tenable of all is the view of Pauli, Altital.
Stud, iii, p. 134 : ' Marcus fur Marticus ; das Marti- ist = lat. morti-
mit alterer Vokalisation.' Marcus appears not to be derived from
Mars, but to represent an old compound name, one of whose mem-
bers was Maro-, and to be formed with the diminutive ending -acus ;
cf. such names as Skt. Deva-ka, Vira-ka, Quna-ka; Gk. °Iirira-Kos,
Hvppa-Kos, Kvva£, etc. The older Latin form was * Mar-ac-os,
which by vowel weakening became * Mar-ic-us, then by syncope, which
occurred especially after a long vowel, Marcus ; cf . optumus < *bpit-
umus, Caldus <C calidus, valde < valide, etc. The same root mar we
find of frequent occurrence in names of other languages. Its force is
' bright,' ' renowned.' In Irish it appears as mdr, mar, as in Glun-mar,
Teacht-mar ; Cymr. Con-mbr, Cat-mbr ; Celt. Sego-mdrus, Viro-marus,
Brogi-marus, Elvio-marus. As shortened, forms we find in Celtic
Maro and Mariccus. We seem also to have the same root, with
unexplained vowel change, in the Gothic mers, which appears in
OHG. as mar in such names as Waldo-mar, Wini-mar, Wolf-mar,
Berht-mar, Fridu-mar ; Mar-oald, Mar-win, Mar-ulf Mar-berht,
Mar-frid ; and in the shortened forms Maro, Goth. Mirila. In
Slavic also we find the same member appearing as miru in the
names Miro-gniew, Miro-dar, Miro-lub, Miro-neha, Miro-slav ; Bole-
mir, Brani-mir, Vladi-mir, Desi-mer, Rado-mir; and in the abbre-
viated names Mir, Mirin, Mirlko, Mirota, Miros. The Celtic Maric-
cus and the Slavic Miriko are therefore the exact equivalents of the
Latin Marcus.

VI. Women's Praenomina.

In the classical literature of Rome women are not designated by
any praenomen. In republican times the commonest form of
designation was the gentile name alone, as Caesennia (Cic. pro Caec.
10), Ampia (Cic. ad Fam. vi, 12, 3), Aquilia (Cic. ad Att. xiv, 13, 5 ;
I7> 3)- When this was insufficient to indicate the individual, the
name of the husband or nearest male relative, or some term of
relationship was added, as Anniae C. Anni senatoris \_flliae], Cic.
Verr. i, 153; Auria fratris uxor, Cic. pro Cluent. 31; Caeciliam
Nepotis filiam, Cic. pro Sex. Rose. 27 ; Cornelia tua [uxor], Cic. ad

160 George Davis Chase.

Fam. v, 6, i ; Lucretiae Trkipitini filiae Conlatini uxori, Cic. de Rep.
ii, 46 ; Porciae meae, Cic. ad Brut, i, 17, 7 ; Postumia Sulpicii [uxor],
Cic. ad Att. xii, 11. Sometimes an adjective from the place of birth
was added, as Numitoria Fregellana, Cic. Phil, iii, 17. These women
seem to have had no other name in either official or familiar use.
Cicero always calls his daughter Tullia, or when he wishes for a more
familiar name, he makes a pet diminutive, Tulliola.

Occasionally we meet with cognomina in the republican period, as
Aurelia Orestilla, Sail. Cat. 15, 2; Aelia Galla, Prop, iv, 12, 38.
Oftener such women are mentioned by their cognomina alone, as Ore-
stilla, Sail. Cat. 35, 36; Cic. ad Fam. viii, 7, 2 (but Aurelia, ix, 22, 4);
Fausta {Cornelia), Cic. ad Att. v, 8, 2 ; Metella, Cic. post Red. in Sen.
37 ; ad Att. xi, 23, 3 ; Cana (Gellia), Cic. ad Alt. xiii, 41, 1 ; Tertulla
{Tertia), Cic. ad Att. xv, 11, 1 ; xiv, 20 (but Tertia, ad Brut, ii, 5, 3;
6, 2 ; ad Fam. xvi, 22, 1). In speaking of Atticus' daughter, Cicero
sometimes calls her Caecilia {ad Att. vi, 2, 10; 4, 3), sometimes Attica
{ad Att. xii, 1,1; xiii, 15, 17 ; 21, 7 ; xvi, 11, 8), and once Atticula
{ad Att. vi, 5, 4). Thus we see from such cognomina as Attica,
Metella, Lepida, etc., that an additional name might readily be made
from the father's cognomen.

Under the empire we find much the same state of affairs, except
that there is a notable increase in the use of cognomina, as Aelia
Paetina, Tac. Ann. xii, 1 ; Aetnilia Lepida, vi, 40 ; Aemilia Musa, ii,
48; Annia Rufilla, iii, 36; Appuleia Varilla, ii, 50; Artoria Flaccilla,
xv, 71 ; Atria Galla, xv, 59 ; Calvia Crispinilla, Hist, i, 73 ; Claudia
Pulchra, Ann. iv, 5 ; Domitia Decidiana, Agric. 6 ; Egnatia Maximilla,
Ann. xv, 71; Pompeia Celerina, Plin. Epist. i, 4 ; Ummidia Quadra-
tilla, vii, 24, 1; Galeria Copiola, Plin. N. H. vii, 158; Lollia Paulina,
ix, 117; and numerous similar names.

From the evidence of the use of names in literature there is little
to lead us to suppose that Roman women ever had praenomina.
There are, however, other kinds of evidence equally important which
we must examine. In the first place, in the original Indo-European
system women were named exactly in the same manner as men, and
any theory which derives Roman praenomina from this system may
fairly look for vestiges of praenomina among the names of women.
Secondly, Varro in his researches has gathered for us some very

The Origin of Roman Praenomina. 161

valuable information. The following passage is of particular interest
(Z. Z. be, 60) : Itaque ibi apparet analogia ac dicitur Terentius vir,
Terentia femina, Terentium genus. In praenominibus ideo non fit
item, quod haec instituta ad usum singularia, quibus discernerentur
nomina gentilicia, ut ab numero Secunda, Tertia, Quarta, in viris ut
Quintus, Sextus, Decimus, sic ab aliis rebus. Cum essent duo
Terentii aut plures, discernendi causa, ut aliquid singulare haberent,
notabant, forsitan ab eo, qui mane natus diceretur, ut is Manius
esset, qui luci, Lucius, qui post patris mortem, Postumus. E quibus
aeque cum item accidisset feminis, proportione ita appellata declina-
rant praenomina mulierum antiqua, Mania, Lucia, Postuma ; videmus
enim Maniam matrem larum dici, Luciam Volaminiam Saliorum
carminibus appellari, Postumam a multis post patris mortem etiam
nunc appellari.

Also the Incertus Auctor of the liber de Praenominibus, who
was a student of Varro, has the following interesting and per-
tinent passage (§ 7): Antiquarum mulierum frequenti in usu prae-
nomina fuerunt, Rutila, Caesellia, Rodacilla, Murrula, Burra a colore
ducta. Ilia praenomina a viris tracta sunt, Gaia, Lucia, Publia,
Numeria. Ceterum Gaia usu super omnes celebrata est. Ferunt
enim Gaiam Caeciliam, Tarquinii Prisci regis uxorem, optimam
lanificam fuisse et ideo institutum ut novae nuptae, ante ianuam
mariti interrogatae quaenam vocarentur, Gaias esse se dicerent.

Although the author of this brief article cites Varro, he relied
mainly on other authorities, as is shown from the fact that his list
of praenomina is quite different from that of Varro, and even in
one place they are in contradiction. The Incert. Auct. mentions
Numeria in his list, but Varro distinctly says, Z. Z. ix, 55, 'sic esse
Marcum, Numerium, at Marcam, at Numeriam non esse.'

For further light we must turn to our collection of Roman inscrip-
tions. These, found in every quarter of the empire, and represent-
ing different periods of Roman history, contain a large percentage
of women's names. Taking CIL. xiv, as a fair sample, we find that,
out of a total of about 5000 names, 187 1 are of women. The names
represent all social classes, but the majority belong to the undis-
tinguished multitude, and would never have survived to our day had
it not been that their owners were fortunate enough to have the fact

1 62 George Davis Cfiase.

of their death recorded. Naturally, imperial times are most fully
represented and show, as the commonest type, gentilicium and
cognomen, of which the latter is either a Latin adjective, as in Aelia
Optata, CIL. ii, 5492, or evidently a derivative from a masculine
cognomen, as in Aelia Ruftna, CIL. ii, 990, or a Greek or foreign
slave name, as in Aemilia Philumene (OiXov /ten?), CIL. ii, 6160.

In seeking for Roman praenomina we must be careful to exclude
those names of southern Italy which are evidently not Roman, but
belong to the system in vogue in Oscan territory, as Vestia Oppia
Atellana, Li v. xxvi, 33; Val. Max. v, 2, 1; Paculla Annia Campana,
Liv. xxxix, 13; Cluvia Facula, Val. Max. v, 2, 1 ; Ennia Naevia,
Suet. Cal. i2.

In this list I would also class the following :

CIL. i, 54. Dindia Macolnia (Dindia as nomen, Orell. 3784).

" i, 149. Maria Selicia.

" i, 1062. Volminia D. 1. Salvia.

" i, 1063. Salvia M. 1. Servia.

" i, 1063. Iulia M. 1. Ammia.

" i, 1082. Pupia L. 1. Statia.

" i, 1 183. Agria Sueia N. f.

" i, 1204. Asula Sex. 1. Salvia.

" i, 1298. Gavia Caesidia.

" i, 1 50 id. Rudia Vergelia.

" xiv, 3134. Maria Fabricia.

" xiv, 4104. Ceisia Loucilia.

" xiv, 1293. Statia Pompeia.

" xiv, 2938. Tutia Marcia.

" ix, 1 17 1. Vibia Licinia Telesilla.

" ix, 6335. Vibia Sullia L. f.

" ix, 3272. Vibia Tetidia L. f.

" ix, 255. Vicilia Titania Procula.

" x, 2717. Maria Caecilia Procilla.

" x, 2783. Novia Fl(avia) Thallusa.

" x, 1 102. Sergia Vibia Maxima.

We occasionally find slave names, or other names that are regu-
larly and commonly used as cognomina and have the character of

The Origin of Roman Praenomina. 163

such, placed before the gentile name. These could never have been
considered real praenomina, but were permitted perhaps through the
ignorance of foreigners, or the disturbance under the empire of the
earlier name system. Examples of these are as follows :

Sen.

Contr. ix,

5, 15. Galla Numisia.

Tac.

Ann. xiii,

45. Sabina Poppaea.

CJL

• i> !S5-

Graeca Vatronia.

a

iv, 1799.

Sabina Sulpicia.

a

v, 449.

Sabina Laevica Mercii f.

a

ix, 3827.

Pampila Anaia P. 1.

a

ii, 1 149.

Galla Blasti f. Servilia Superati.

ii

ix, 193 1 add. Hispania Pomponia.

tt

ii, 2356.

Nigrina Sulpicia.

a

i, 1301.

Rutila Fukinia.

a

ix, 4666.

Rutila Iapriena T. f.

a

ix, 4387.

[Rjutila Erefria.

tt

v, 6550.

Rufa Gall M. f.

tt

ii, 3470.

Caesilla T. f. Cornelia.

it

i, 1461.

Grata Plotia.

it

v, 6612.

Optata Clod[ra].

a

v, 4676.

Optata Mulvia.

it

v, 454-

Hospita Petronia P. f.

it

v, 7306.

Esiata Oppia.

it

ix, 3248.

Saluta Obellia.

a

viii, 9386.

Rogata Fabricia Procli f.

a

iii, 1612.

Afiicta Caesia.

a

X, IOOI.

Apta Buccia (but 1002 Bucia Apta).

a

xii, 35 x9-

Donata Cirratia.

a

x, 1299.

Modesta Fisia Rufina.

a

x, 55I6-

Rufa Rasinia.

a

iii, 3038.

Avita Aquillia L. f.

a

iii, 3!5r-

Avita Nigidia Volsun. f.

tt

iii, 3038.

Avita Suioca Vesclevesis.

it

v, 2606.

Urbana Claudia.

a

ii, 945-

Severa Mania L. f.

a

ix, 1228.

Ingenua Babria.

164 George Davis Chase.

CIL

• v, 3772.

Prisca Terentia Q. f. Felicil

ti

v, 8156.

Bona Titacia.

it

v, 3070.

Agraeca Vosinia.

it

viii, 5013.

Felicia Antonia.

a

ix, 4933-

Posilla Senenia Quart, f.

a

ix, 4763b.

Stella Apollonia.

tt

xii, 682 a.

Silvana Patricia.

tt

xii, 4193.

Saturnina S[eve]ria.

a

xii, 3945-

Primigenia Aurelia.

a

ii, 2874.

Titulla Ticconia.

a

vii, 58.

Succ[essa] Petronia.

a

ii, 1788.

Sura Cercia.

a

in, 5265.

Vetulla Bucia Urbani f.

a

iii, 5621.

Romana Argentonia.

a

v, 5387-

Suadilla Bellia Cripponis f.

ti

v, 7349-

Vera Blaionia.

tt

v, 61 2 1.

Verina Boiemia.

ti

v, 3536.

Pupa Cassia M. f.

tt

v, 4109.

Pusilla Clodia.

tt

v, 5°33-

Verecunda Fundania.

a

v, 449.

Marcella Laepoca.

tt

v, 449.

Ternila Laevica Regiliae f.

tt

ix, 65.

Acerratina Salvia.

tt

v, 4126.

Seneca Magia Magi f.

a

v, 4894.

Mesa Manilia Sev ....

ti

v, 7306.

Gaudilla Oppia.

a

ix, 1527.

Verna Rufria.

a

ii, 3453-

Philocale Fadenia.

tt

ii, 2297.

Nice Numisia.

tt

ii, 6207.

Stelea Atilia Fortunata.

tt

xiv, 3105.

Gemna Cordia.

a

ix, 6413.

Antiopa Ania.

tt

ix, 321 1.

Danais Aufidia.

tt

v, 37 19-

Thedista Quintia.

tt

v, 7641.

Enica Comiogia Nevi f.

ti

v, 7856.

Mocea Eunania.

it

i, 918.

Dercina Nanalaria.

The Origin of Roman Praenomina.

165

CIL. i, 982. Himinis Terentia.

i, 1 2 19. Acumis Volusia.

ix, 1422. Erotice Vircinia.

ix, 378. [Me]rope Faenia.

ix, 378. [Theojdote Galbia.

i, 1356. Philomena Satria.

ix, 5486. Auge Obilia.

ix, 3200. Agathemis Pulfidia.

x, 4423. Anthedonium Volusia,

xii, 3424. Lycoris Anicia.

xii, 809. Hilara Appaia.

xii, 3945. Primigenia Aurelia.

xii> 3595- Helene Gaetulia.
xii, 4676add. Calliste Martia.

xii, 4465. Cloe Valeria.

In the inscriptions from Etruria are found the Etruscan names
Velisa or Valisa {CIL. i, 1350; 1365) and Thania or Thannia(CIL.

i. 1347; ^63 ; 137°; I37I; !373; 1375)- The latter also appears
in a Faliscan inscription as a cognomen, Zvetaieff, Inscrip. Ital. 56.

Fausta is found more frequently than the others which we have
gathered, but should be classed with them :

CIL. viii, 142 1. Fausta Servia.

" v, 2486. Fausta Lartia.

" ix, 3100. Fausta Vibia.

" x. 8353- [F]aus[t]a Iulia.

No one would claim, I think, that any of the names in the above
lists, though written first in some instances, were ever really praeno-
mina. We come to others, however, about which, owing to their
difference in use, a doubt might more properly be entertained.
These are a limited number of names which seem to be used either
first or last with equal freedom. We will consider them separately.

Paula we find preceding the gentile name in the following instances :

Cic. ad Att. viii, 7, 2. Paula Valeria.
CIL. i, 39. [PJaulla Cornelia Cn. f. Hispalli.
" i, 177. Pola Livia.

1 66 George Davis Chase.

CIL. i, 952. Paulla Salvia.

" i, 1034. Polla Caecilia Spuri f.

" i, 1 155. Paul Toutia M. f.

" i, 1303. Pola Aponia.

" i, 1313. Pola Abelese.

" xiv, 3453. Pola Suestidia.

" ii, 4623. Paulla Aemilia.

" ii, 4363. Paulla Fulvia.

" v) 7719- Polla Lebronia Terti f.

" ix, 4078. Pol[l]a Cosidia.

" ix, 4239. Paulla Lacutalana Q. f.

" ix, 4142. Pola Runtia L. f.

" x, 5148. Polla Betuedia.

" x, 6166. Polla Minculeia M. f.

" x, 2855. Polla Victoria.

Of these eighteen cases, the first eight are certainly from republi-
can times, and some of the others may be. The cases where Paula
follows the gentile name are considerably more numerous, — too
numerous to give here, — but none of them are clearly republican.

An interesting pair are Maior and Minor, which are used in in-
scriptions even of an early period, in the position both of praenomen
and cognomen, but in literature often have no more force than
descriptive adjectives. Examples of the latter use are as follows:

Fronto, ii, 13, Gratia minor, Gratia maior (mother and daughter).

Suet. Aug. 4, Octavia minor, Octavia maior (sisters).

Suet. Nero, 5, Antonia maior; Suet. Cal. 1, minor Antonia.

Liv. vi, 34, minor Fabia.

Sen. frag. 76, Haase, Marcella maior ; Porcia minor.

Cf. Cato maior, in Cicero.

In inscriptions we find them in use as follows:

CIL. i, 78. Minor Ania C. f.

" i, 97. Mino Coponia Artorom ....

" i, 153. Min. Tutia.

" xiv, 31 1 1. Mino Cumia L. f.

" xiv, 3166. Mino Mamia Tib. f.

The Origin of Roman Praenomina. 167

CIL. xiv, 3167. Mino Matlia.

" i, to8. Maio Fabricia.

" i, 136. Maio Orcevia M. f.

" xiv, 3057. Maio Anicia.

" xiv, 3301. Maio Fortun ....

" xiv, 3284. Maio Tutia Q. f.

" xiv, 3299. Maio Volentilia.

" xiv, 3974. Herennia L. f. Merula minor.

" xiv, 3237. Samiaria M. f. Minor Q.

" xiv, 3973. Herennia L. f. Merula maior.

" xiv, 225. [Sext]ilia Maior.

" i, 928. Otacilia Maior (?).

" ii, 651. Caecilia Q. f. Maior.

It is to be noted that of this list all but the last one are from
Latium. While twelve have the name preceding the gentilicium, six
only have it following, and of these CIL. i, 928, is a doubtful read-
ing, and in CIL. xiv, 3973 and 3974, the maior and minor appear to
be simply distinguishing adjectives and not personal names.

Maxima is almost always first, as follows :

CIL. i, 1256. Maxsuma Sadria S. f.

" ix, 5058. Maxutna Caesedia T. f.

" ix, 5803. Maxima Nasica Cn. f.

" i, 1434. Maxuma Aimilia.

" v, 3180. Maxima Lucilia.

" xii, 3981. Maxsuma Pompeia.

" v, 2595. Catulla Q. f. Maxima.

" x, 1 102. Sergia Vibia Maxima.

Derivative forms appear in the two following inscriptions (else-
where following the gentile) :

CIL. iii, 3989. Maximana Aemilia.
" iii, 2615. Maximilla Poppia.

We now come to those names which, according to the evidence
of the Romans themselves, were considered praenomena. The
first one of these to take up is Postuma. Let us recall the words of

1 68 George Davis Chase.

Varro, L. L. ix, 60 : ' qui post patris mortem (natus), Postumus dicere-
tur. E quibus aeque cum item accidisset feminis, . . . declinarant
praenomina mulierum antiqua, Mania, Lucia, Postuma ; videmus enim
. . . Postumam a multis post patris mortem etiam nunc appellari.'
Plut. Sulla 37 bears similar testimony: 'H yap OvaWtpCa /jlcto. ttjv
TcXevrr/v avrov OvyaTpiov airtKvrfcrtv, o THocrrovpav IkoXovv tous yap
varepov Trjs tS>v iraTfptov TeAevrijs ya>op.ivov%, ovr<a Voip.atoi. irpoaayo-
pevova-iv. The examples all show Postuma as a true cognomen :

CIL. xiv, 2418. Herennia Postuma.

" ii, 3740. Marcia P. f. Postuma Messenia Lucilla Aemilia.

" ii, 1674. Anicia Sex. f. Postuma.

" ii, 2654. Domitia Postuma.

" iii, 3133. Kapia Postuma.

It must be admitted that Varro's statement finds very little sup-
port, and that his words are the only evidence that Postuma was ever
a praenomen.

The numerals are next in order. These require our special atten-
tion because of their frequency of use as praenomina and their
antiquity as names, as well as because of the evidence of Varro. We
have said that in republican times women were most commonly
designated by the gentile name alone. For the eldest daughter this
did very well, but where there were younger ones, some means of
distinguishing were necessary. We catch glimpses of cases where
the numerals were used for this purpose, and it is likely that the
practice was more frequent than appears from the evidence.

L. Aemilius Paullus, the hero of Pydna, had two daughters by his
first wife, one of whom married Cato, the other Tubero (Plut. Aem.
5). Upon his second election to the consulship, in 168 B.C., when
his two sons by his second wife were respectively eleven and thirteen
years of age, we are told by Cicero (Div. i, 103) that his daughter
Tertia was a little girl : L. Paulus consul iterum, cum ei bellum ut
cum rege Perse gereret obtigisset, ut ea ipsa die domum ad ves-
perum rediit, filiolam suam Tertiam, quae turn erat admodum parva,
osculans animum advertit tristiculam. "Quid est," inquit, "mea
Tertia ? quid tristis ? " " Mi pater," inquit, " Persa periit." Turn

The Origin of Roman Praenomina. 169

ille artius puellam complexus, "accipio," inquit, "mea filia, omen."
Erat autem mortuus catellus eo nomine. From her age, therefore,
and what we know of Paullus' family, we may be certain that she
was actually his third daughter in order. She is again spoken of as
Tertia in Cic. Div. ii, 40, and in Plut. Aem. 10, where the same
story is cited from Cicero. Still another mention is found in Val.
Max. vi, 7, 1, Tertia Aemilia, Africani prioris uxor, mater Corneliae
Gracchorum.

Another Tertia was the half-sister of M. Brutus and the wife of C.
Cassius. She is variously spoken of as Tertia (Suet. lul. 50 ; Cic.
ad. Brut, ii, 5, 3 and 6, 2), Tertulla (Suet. Jul. 50 ; Cic. ad. Att. xv,
n, 1 ; xiv, 20, 2), Iunia Tertia (Macrob. Sat. ii, 2, 5), and Junta (Tac.
Ann. iii, 76 : Iunia. . . . supremum diem explevit Catone avunculo
genita, C. Cassi uxor, M. Bruti soror). She had an elder sister who
is called Junia simply in Cic. ad. Att. xiv, 8 ; Veil. Pater, ii, 88.

The eldest daughter of Q. Mucius Scaevola is called Mucia in
Cic. Brut. 211. A younger daughter is spoken of as follows by
Asconius {in Scaur. 19) : nam Tertiam, Scaevolae filiam .... duxe-
rat. A few lines below she is spoken of as Mucia. She is also
called Mucia elsewhere (Cic. ad. Att. i, 12, 3 ; ad. Fam. v, 2, 6 ;
Suet. Caes. 50 ; Plut. Pomp. 42).

The tribune P. Clodius had three sisters, according to Cicero {ad.
Fam.i, 9, 15): ilia furia muliebrium religion um {sc. Clodius) qui non
pluris fecerat Bonam deam quam tris sorores. There is a statement in
Varro which seems at first sight to contradict this. Appius Claudius,
the brother of P. Clodius, is made to say (Var. R. R. iii, 16, 2) : 'nam
cum pauper cum duobus fratribus et duabus sororibus essem relictus,
quarum alteram sine dote dedi Lucullo.' The contradiction, how-
ever, is only apparent, as it may well have been the case that only
two sisters were left dependent on Appius ; the eldest may have been
settled before the father's death. We learn more particularly about
these sisters from Plutarch, Cic. 29, AtwovAAos Be ko.1 QepairaiviSas
irap«X£V> <*S (TvyyevoiTO ry vecoTaVg r<Sv aSe\<f>wv 6 KAuJSios, ore AevKovXXm
o-vvioku. U0XX1) 8* rjv Sofa kcu reus aAAais Svcrtv dScA^ais irXrjO-idfciv
rbv KAwSiov, <5v TepTi'av /xev Map/aos 6 'Prj$, KAwoW oe Mo-cMos 6
KtXep elx^v, fjv KovaBpavrlav IkoKow. There can be no doubt that the
youngest sister married Lucullus. This is plain both from the

170 George Davis Chase.

statement of Plutarch and from the fact, according to Varro, that
she was married without dowry, which would be most likely to be the
case with the youngest. We have next to decide between Tertia and
Quadrantaria as to the order of their birth. The eldest daughter
would be Clodia par excellence, and Quadrantaria is so styled
in Plutarch. She had no other name than Clodia, except the
abusive epithets which Cicero bestowed. Plutarch's language decides
this point. When he speaks of the two as Clodia and Tertia his
meaning is as significant as when in English we say ' Miss Wilson '
and 'Miss Alice,' referring to two sisters. The order must have
been, therefore, (1) Clodia (Quadrantaria), (2) Tertia,<($) the wife
of Lucullus. It does not seem probable that Tertia could ever have
meant the third child, for the number of sons in a family would not
have affected the designation of the daughters. Least of all would
she be called Tertia if she was the eldest daughter, but had two
older brothers ; it was not only the custom but an honor to be
known only by the gentile name. In our particular case Tertia
Clodia was probably the fourth child, since her brothers Appius and
Gaius appear to have been older. But that she was the third
daughter and still had only one older sister is not so paradoxical as
it at first seems. When we take into account the rate of mortality
among young children, the chances are at least even that there was
a second daughter who died as a child. On the whole, the evidence
seems convincing that Tertia in republican times meant the third
daughter.

A mistress of Verres is mentioned five times in Cicero ( Verr. iii,
34 ; v, 12, 16) by no other name than Tertia.

As to the position of Tertia with respect to the gentile name,
we may notice that, in the only cases where they occur together,
Valerius Maximus wrote Tertia Aemilia, and Macrobius, Junta
Tertia. In the time of "Valerius the order was already beginning
to be settled, and Macrobius is very poor authority indeed. The
fact which seems to be pointed to, that the numerals were always
descriptive, is an important one, and denotes a marked distinction
from the use of the numerals among men. There Quintus could as
readily be used for the firstborn as for the fifth, — indeed more so if
the father's name happened to be Quintus (see above, p. 151). Now

The Origin of Roman Praenotnina. 171

a praenomen must be defined primarily as the distinguishing name
of the individual, which might be chosen arbitrarily, and had little
more significance than Charles or Henry with us. The fact that a
name was in any way descriptive of the individual stamped it as
a cognomen, and such we must consider the numerals in the case
of women essentially to have been. Varro's classification of them
as praenomina doubtless arose from his desire to reduce them to
the same category as Quintus, Sextus, etc., where, indeed, they once
properly belonged. They afterwards went through the same evolu-
tion that Quintus, etc., had gone through, so that a hundred years
after Cicero's time Tertia, Tertina, or Tertulla no longer meant the
third daughter, and such names as Primula P.f. Secundina (CIL. xii,
2761) were possible. But by this time the distinction between prae-
nomen and cognomen was so much broken down that there was no
clear consciousness of their being either.

We learn from tradition that Romulus named his eldest daughter
Prima: Plut. Rom. 14, kclL yevecrOai km iraiSas avr<j> (Romulus), /xiav
fiiv Ovyartpa llpifiav, rfj rd£ei tijs ycvttrceos ovtu> TrpovayopevOeurav.
Prima occurs in inscriptions as follows :

CIL. i, ioio. Prima Pompeia.
" v, 2608. Prima Coelia Mataronis.
" v, 2805. Prima Minucia.

Of Prima as cognomen we find fifty-four instances in CIL. i-iv
and xiv, and a proportionate number throughout. In Plin. Epist.
Tra. 60 occurs the name Furia Prima. It will be remembered that
Varro admitted as praenomina the numerals Secunda, Tertia, and
Quarta only. The derivative forms of the numerals, as Primula,
Secundina, Tertiola, Tertulla, Quarlilla, Quintilla, Quintina, etc.,
occur so constantly as cognomina that we need mention them only
in the few instances where they are found as first names, as follows :

CIL. xii, 2761. Primula P. f. Secundina.
" xii, 5219. Primula Verria.
" xii, 2827. Secundina Cominia.
" ix, 4351. Tertulla Albia Pamphili 1.
" viii, 877 add. Quartina Pupenia.

172

George Davis Chase.

For Secunda we find in two inscriptions the additional evidence of
its conscious use as a praenomen that it was abbreviated to Sec.
The examples of Secunda are as follows :

CIL. v, 996.

" ix, 3862.

" ii, 614.

" v, 2120.

" v, 3488.

" v, 3783.

" v, 2327.

" v, 8877.

« v, 5177.

" v, 3794.

" x, 5277.

" x, 3818.

Sec. Petronia C. f. N. . . .
Seq. Frendesia P. f.
Secunda Herennia.
Secunda Accia.
Secunda Annia L. f.
Secunda Baebia C. f.
Secunda Cammica Siponis f.
Secunda Octavia C. f.
Secunda Pompeia.
Secunda Valeria M. f.
Secunda Lusia L. f.
[Se]q[u]nda Solania L. f.

It is significant that eight of these twelve inscriptions come from
Gallia Cisalpina. A local fashion appears to have swelled the num-
ber considerably. In CIL. i-iv and xiv alone, we find seventy cases
of Secunda used as a cognomen. The proportion agrees well with
Prima.

Tertia comes first as follows :

1. Tertia Aemilia.
Tertia Basilia.
Tertia Sapiena C. 1.
Tert Coriaria.
Trtia Sauna.
Tertia Aniavia.
Tertia Apicia.
Tertia Cominia.
Tertia Dometia Maconi f.
Tertia Petronia M. f.
Tertia Vippia Vippi f.
Tertia Boel[ia] Salvia.
Tertia Turpedia.
Tertia Rubria [Tjerti f.

Val.

Max. vi, 7,

CIL

i, 1099.

u

i, 1298.

It

xiv, 3107.

a

xiv, 3251.

a

v, 2370.

tl

v, 2563.

a

v, 7385-

it

v, 6931.

it

v, 7385-

a

v, 7961.

a

ix, 4375-

tt

ix, 5169.

u

ix, 4386.

The Origin of Roman Praenomina. 173

Here again we notice that out of thirteen nine are from northern
Italy. It may be that these represent an older Roman fashion, a
' patavinitas ' that had gone out of use in the city itself. In CIL.
i-iv and xiv we find twenty-eight cases of Tertia used as a cognomen.

Of Quarta there are the following examples :

Liv. xl, 37. Quarta Hostilia.

CIL. i, 1306. Quarta Senenia C. 1.

" xiv, 3283. Quorta Tondia.

" v> 3772. Quarta Terentia.

" ix, 4718. Quarta Decia L. 1.

" x, 3817. Quarta Confleia M. 1.

As cognomen Quarta occurs in CIL. i-iv and xiv seventeen times,
and frequently elsewhere. In Latium, however, only one example
is found, — CIL. xiv, 287, Acilia Quar.

With respect to Quinta the case is somewhat more complicated.
The chaste Claudia of 204 B.C. appears in Cic. pro Cael. 34 as
Quinta ilia Claudia, in de Harus. resp. 27 as Q. Claudia, but with
the variant readings of Quincta in cod. P1, and Quinta in P2G. Her
name is, however, handed down as Claudia Quinta in Liv. xxix, 14,
Tac. Ann. iv, 64, and Ovid, Fasti, iv, 305. In Val. Max. i, 8, 11,
the MS. is corrupt : Kempf in his earlier edition restores Q. Claudiae,
in his later one Quintae Claudiae. Of the inscriptions in which the
name occurs the majority come from Africa, and in most of these
Quinta is abbreviated to Q. The examples are as follows :

CIL. viii, 6227. Q. Iulia Gelulici f.

" viii, 3755. Q. Iulia Urbana.

" viii, 9216. Q. Martia.

" viii, 6607. Q. Sitia Novella.

" viii, 565. Q. Tupidia Ianuaria.

" v> 3572. Q. Cornelia C . . . . Rufialis.

" viii, 6863. Quinta Aemilia.

" viii, 5300. Quinta Caecilia.

" viii, 8356. Quinta Iulia.

" xiv, 1649. Quinta Sulpicia.

" ii, 3680. Quinta Caecilia Norisi.

174 George Davis Chase.

CIL. ii, 2945. Quinta Fabia.
" xii, 1385. [Q]uinta Centon(ia ?).

The habit of abbreviating to Q. undoubtedly arose from the
abbreviation of the masculine praenomen Quintus. We shall see
that all the women's names which had corresponding masculine
praenomina were regularly so abbreviated. In CIL. i-iv and xiv
we find twenty-six cases of Quinta as cognomen.

Of other numerals we find very scanty examples. We seem to
have Sexia in CIL. x, 5354, Sex. Cornelia Panteris. In CIL. v,
6455, Oct. Valeria Vera, we may suppose either Octava or Octavia;
perhaps the latter is the more probable. As examples of cognomina
we may cite as follows :

CIL. iii, 2412. Lalia Sexta.

" iii, 3187. Ostoria Sexta.

" xiv, 501.2. Aebutia Septima.

" xiv, 2338. Casperia Septima.

" ii, 2592. Carisia Nona.

We have now reached a short final group, the feminine forms of
real masculine praenomina, as Gaia, Lucia, Publia, Mania. For
Gaia there is considerable evidence from literature, as follows :
Incert. Auct. de Praen. 7, quoted on p. 161 ; Paul, ex lest. 95, Gaia
Caecilia appellata est, ut Romam venit, quae antea Tanaquil vocitata
erat, uxor Tarquinii Prisci regis Romanorum, quae tantae probitatis
fuit ut id nomen ominis boni causa frequentent nubentes, quam
summam asseverant lanificam fuisse ; Plut. Quaest. Rom. 30, Aia ti
r-qv vv/J.<t>i]V eicrdyovTK Aeyeiv KtXevovo-iv, Oirov cru Tat'os, iyl> Ta'Ca.
Horepov, cooTrep eiri pijrois ev$vs ela-euri T<5 KOivoweiv airavrtav Kal avvdp-
X«v," Kal to p.ev 8rj\ovp.€v6v eoriv, "Oirov crv Kvpios Kal oiKoSeoTroViys, Kal
ty<o KvpCa Kal oixoSecnroiya • Tots S ' ovo/jxuti tovtois aXAcos K€)(prjvTai
Koivot'S 0Z0-LV, tocrirep 01 vop.iKol Taiov Sifiov Kal Aovkiov Tmov, Kal ot
<j>i\6<jo<t>ot AiWa Kal ®ia>va TrapaXdp.fiavovcnv ; *H 81a Tatav KaiKiXiav
KaXijv Kal dyaOrjv yvvaiKa, tu>v TapKWtov muoW evl o-vvoiKrjo-acrav, i;s iv
t<5 tov SayCTOV lep<S xaXKOvs avSpias eaTij/cev ; Ikoto St rrdkai Kal o-av-
8a\ta Kal arpaKTOL, to p.ev oiKovpias auTjJs, to 8i, evtpytuK o-v/j.fio\.ov ;
Plin. N. H. viii, 194, Lanam in colo et fuso Tanaquilis, quae eadem

The Origin of Roman Praenomina. 175

Gaia Caecilia vocata est, in templo Sanci durasse prodente se auctor
est M. Varro, factamque ab ea togam regiam undulatam in aede
Fortunae qua Ser. Tullius fuerat usus; Paul, ex Fest. p. 224, prae-
nominibus feminas esse appellatas testimonio sunt Caecilia et
Taracia quae ambae Gaiae solitae sint appellari, pari modo Lucia
et Titia ; Cic. pro L. Murena, 27, In omni denique iure civili aequi-
tatem reliquerunt, verba ipsa tenuerunt, ut, quia in alicuius libris
exempli causa id nomen invenerant, putarunt omnes mulieres quae
coemptionem facerent, Gaias vocari ; Quint, i, 7, 28, nam et Gaius
C littera significatur, quae inversa mulierem declarat, quia tarn Gaias
esse vocitatas quam Gaios etiam ex nuptialibus sacris apparet ; Veil.
Long. 2218, C inversum quo Gaia significatur; quod notae genus
videmus in monumentis, cum quis libertus mulieris ostenditur :
Gaias enim generaliter a specie omnes mulieres accipere voluerunt.
The account is fairly consistent and we are inclined to believe it in
the main ; namely, that Gaia was an old and very common prae-
nomen, and came to be used, just as some people call every boy
Johnny, to stand for any one of a class, that is, of the class of
women. The passage from Cicero partially explains the marriage
formula. Gaius stood for any man's praenomen and Gaia for any
woman's. In the service we may understand that the real names
of the contracting parties were substituted. It is interesting to
compare the marriage formula in Skt. Afval&yana Grihyasutra, i, 7,
where the groom says,

' amo 'ham asmi, sa tuam ;
sa tvam asi, amo aham,'

in which personal pronouns are used without the names. That
C inversum was used to designate the libertus of a woman is
amply attested by the evidence of the monuments. In Gell. vii, 7, 1,
we find the name Gaia Taracia, but in Plin. N. H. xxxiv, 25, the
same name appears as Taracia Gaia ; cf. Paul, ex Fest. p. 224. Val.
Max. viii, 3, 2, speaks of C. Afrania, who lived in Cicero's time, and
whose name became a reproach among women for public impudence.
Inscriptions furnish us with the following instances of Gaia :

CIL. v, 137. C. Basilia Crispina.
" v> 7959- C. Valeria Candidi[lla ?].

176 George Davis Chase.

CIL. viii, 3391. C. Annia Maximina.

" viii, 300. Gaia Iul[ia].

" viii, 10520. C. Sulpicia Ro[g]ata.

" viii, 3664. Gaia Iulia C. Iulii Celeris f.

In the two following the readings are doubtful :

CIL. iii, 4721. C. Iulia Iuliana (?).
" ix, 2712. C. Attia Sabina (?).

In CIL. viii, 3348, we have

D M

CAELI MAC
RINA VIXIT, etc.,

which Renier reads C. Aelia Macrina. Mommsen, however, doubts
this interpretation, and it seems very questionable. In the inscrip-
tion given by Zvet. 35, CIA-PACIA have been conjectured by
Deecke (Zvet. App.) to stand for C(A)IA PACIA.

Lucia is mentioned as a praenomen by both Varro and the Incert.
Auct., and appears in the following inscriptions :

CIL. i, 1357. L. Sentia Sex. f.

" v, 2209. L. Barbia Progenita.

" viii, 3869. L. Antestia Saturnina.

" viii, 7578. L. Manlia Honorata.

" xii, 397. L. Caecilia L. f. Donata.

" xii, 706. L. Statia Firma.

" i, 194. [Lujcia Pacia (cf. Zvet. 35).

" vi, 1398. L. Baebia Sallustia Crescentilla C. f.

" vi, 1516. L. Septimia Patabiniana Balbilla Tyria, etc.

" v, 950. Lucia Vitellia q. et Senecilla.
Ephem. Epig. v, 1358. Lucia Paccia Valeria Saturninae f.

" i88i,p.223. AOYKIATTOMTTONIA MEAiTINH, etc.
Bullet. Arch. Comun. 1880, p. 24. Lucia Licinia Urbana.
Grut. 447, 3. L. Peducea Iuliana.

The two following are uncertain :

CIL. iii, 42 1 1. L. Septimia Severa.
" xii, 5093. L. Rinnia Primae L. Aucta.

The Origin of Roman Praenomina. 177

Lucia is also not infrequently found as a cognomen, as in the
following inscriptions :

CIL. ii, 3896. Cornel. T. f. Lucia.

" iii, 5680. Barbia Lucia.

" iii, 5947. Gessatia Lucia.

" iii, 1250. Herennia Lucia.

" iii, 4281. Modiasia Lucia.

" iii, 134. Petilia Lucia.

" iii, 1 132. Statilia Lucia.

It is to be noticed that these examples are almost entirely con-
fined to the eastern provinces.

For Publia, besides the statement of the Incert. Auct, we have
the following inscriptional evidence :

CIL. i, 156. P. Vebidia Q. f. Numa.

" ii, 585. P. Valeria Maximina (Momm. Pompeia).

" ix, 3048. P. Aelia Visivilia.

" iii, 1 182. Publia Aelia Iuliana Marcella.

" xii, 5158. Publia Martia.

" xii, 3848. Publia Pompeia.

The reading is uncertain in CIL. iii, 5780, Publia Ceionia Vin-
delica. In CIL. iii, 1184, P- AEL- PROCLA, Mommsen fears that
the whole inscription has been corrupted by conjectures. Publia
appears at least once as a cognomen, CIL. iii, 1249, Viria Publia.

Varro has already told us that the praenomen Mania is derived
from Manius. For the antiquity of the name we have his statement
that the mother of the Lares was called Mania. We find also the
same statement at greater length in Macrob. Sat. i, 7, 34. Cf.
also Paul ex F. p. 144, Maniae turpes deformesque personae ; also
Orelli, 2025, MANIAE DEAE. Manius was regularly abbreviated
by the five-stroke AA/ ; so we find in the very old inscription from
Picenum, CIL. i, 177,

MATRE

MATVTA AA/-CVRIA

DONO-DIIDRO POU-UVIA

MATRONA DEDA

1 78

George Davis Chase.

In Wilmanns, 2675, SEPTIMIA AA/ -LIB- MYRfS, we may
understand either Manius or Mania. For further evidence we have
only the two following doubtful cases :

CIL. i, 867. Mania (?) Fabricia.
" xiv, 1793. M. Ulpia Elpidus (?).

It has already been stated that C inversum was used to distin-
guish the libertus of a woman. In a number of curious cases from
Spain and Gallia Narbonensis, an inverted M seems to have the
same force :

CIL. ii, 558. M • HELVIVS • 1AI • LIB.

ii, 558. MALLIA-IAI- LIB- GALLIA.

" ii, 1449. Mummia W 1. Fortunata.

" ii, 2359. Sulpicia W 1. Callirhoe Superati.

" xii, 4719. Corania W [1], Hymnis.

For further instances of women's praenomina the following in-
scriptions may be noticed :

CIL. i, 84. N. Atilia P. f. (Old Praeneste).
V. Volusia Rufa Q. 1.
Sp. Cassia Quintulli f.
Ap. Aurelia Aur. f. Luperialla, etc.
Heria Calpurnia.

A. Ignia Crispina. (Mommsen thinks this an
interpolation.)

In CIL. xii, 684 T. [I]ulia Valentina (so read by all but Dumont),
the T may stand for Tita. Some additional evidence for the exist-
ence of a praenomen Tita is furnished by the passage from Paul, ex
Fest. p. 224, praenominibus feminas esse appellatas testimonio sunt
Caecilia et Taracia, quae ambae Gaiae solitae sint appellari, pari
modo Lucia et Titia. Tita we could understand as the feminine of
Titus ; Titia is quite incomprehensible. It is most probable there-
fore that a corruption exists in the text of Paulus and that we
should read Tita for Titia.

i,

84.

i,

1102.

xii,

4143-

",

3372-

ii,

1150.

ix,

1272.

The Origin of Roman Praenomina. ijg

The oldest inscriptions from Picenum, CIL. i, 167-180, contain
four names of women, each with a praenomen, Cesula Atilia, M/.
Curia, Pola Livia, .... Nomelia. On these Mommsen says,
' quarum nullam praenomine carere inter indicia est remotissimae
vetustatis.' We prefer to look upon it as a local matter rather than
one of time, and to believe it in keeping with the Italic custom out-
side of Rome of giving women praenomina equally with men. In
two Faliscan inscriptions we find very similar names: Zvet. 58,
Cesula Tiperilia Te. f ; 59, Pola Marcia S. The great majority of
women's names throughout Zvetaieff have praenomina.

It seems to me that names such as Rhea Silvia, and Acca Laren-
tia from the early mythical history of Rome cannot be cited with
any considerable weight in any direction, because their origin and
relation to the historical system of names is entirely obscure.

We have now briefly reviewed the direct evidence for feminine
praenomina among the Romans. For the names from colors, —
Rutila, Caesellia, Rodacilla, Murrula, Burra, — cited by the Incert.
Auct., we have found no evidence whatever, unless Caesillia is a cor-
rupt reading for Caesulla. For the other names cited in the same
place and for those in Varro we found sufficient evidence to lead us
to the following conclusion, which is not absolutely proved, but
seems on the whole the most probable : namely, that women had in
early Roman times a set of praenomina corresponding to those
which we find in use among men ; that these fell into disuse and in
classical times had preserved only vestiges in provincial or vulgar
practice, or in such special uses as the C inversum for liberti of
women and the bridal formula ; that in the general downfall of the
Roman system which occurred in post-classical times, these early
praenomina rose continually, here and there, from the lower social
orders or from provincial localities, but at the same time became undis-
tinguishably mixed with names that were purely and simply cogno-
mina ; and that this confusion now attracted the cognomen to the front,
now relegated the praenomen to a position after the gentilicium.

Paulla, though in origin a cognomen and oftenest so used, came
to be frequently understood and used as a praenomen.

Maior and Minor were hardly more than descriptive adjectives,
as in English Senior and Junior, denoting the relation of wife and

180 George Davis Chase.

daughter, or elder and younger sister. It is doubtful if they were
ever considered really part of the name.

Fostutna seems always to have been a cognomen, Varro to the
contrary notwithstanding.

Of the numerals, Prima was little used for lack of demand ;
Secunda, Tertia, Quarta, and Quinta were used with the force and
most frequently in the position of cognomina, although doubtless
through the influence of the position of Quintus, Sextus, etc., they
were frequently put first.

Survivals of real praenomina scarcely extend beyond the three
names Gaia, Lucia, and Publia, and perhaps Mania, but the
occurrences of these are at the period when any sharp distinction
between praenomen and cognomen was obliterated from the sense of
the people. We see, therefore, that the common statement that
Roman women had no praenomina is both true and untrue. Accord-
ing to the correct use of Cicero's time they had nothing that could
be properly so called ; but in the provinces and among the lower
classes praenomina still survived, ready to come to the surface again
in the social upheavals of imperial times.

VII. Conclusion.

Our work has already outgrown its intended limits, but we must
still linger a moment to state briefly some of the conclusions to
which in our investigation of the facts about Roman names we have
been led. Beginning with the kind of name which is represented by
Demosthenes as a type, and, as we are warranted in doing, assuming
this to be the original form of Indo-European name, we have already
observed that the first important step in the development of the sys-
tem was the shortening process by which Demosthenes became Demon.
In comparison with the related languages, the Italic dialects are
noticeable for their avoidance of compounds of two denominative
members or of a denominative and a verbal member. We are not
surprised, therefore, to find the same tendency so far exerting itself
in the personal names as early to have reduced the compound names
to the form of a single member. At the same time, when we con-
sider that Lucius may be the representative of an unlimited number

The Origin of Roman Praenomina. 181

of compounds having luco- as either first or second member, we see
how the abbreviating process tended to reduce materially the avail-
able number of given names. The number was still more diminished
by the small size and exclusive character of the early Roman settle-
ment. In any community it is the general tendency for the available
number of given names to become very small. In Christian coun-
tries the number is often restricted to the list in the church calendar ;
but in any district the list in common use will be found to have
reduced itself to a small handful. This principle of restriction is so
well recognized in Germany that it has given rise to the Low German
proverb, ' Hinnik un Jan het 'te maisse Mann.'

The possibility of such restriction arises as soon as the individual
assumes, or is given, other additional names as a means of designa-
tion ; and this touches the keynote of the second stage of develop-
ment. Before the separation of the Italic branch from the Greek, a
new kind of name had begun to make its appearance : the nick-
name, often a ' Schimpfname,' and always implying individual
description or criticism, began to be used. In Greek we meet with
such names only sparingly, asIIAaTw, Pdvylikos, etc.; but in Latin
the development grew to be the basis of the widely extended system
of cognomina. The character of these has been already sufficiently
discussed. They must originally have been acquired during the life-
time of the individual, who of course had already received a regu-
larly assigned name. We are therefore forced to believe that the
cognomen came into use as a second name, and at a period prior to
the development of the gentilicium. The latter deduction appears at
first sight to be forced, and not always to be borne out by the facts.
For if the cognomen developed before the gentile we might expect to
find its employment more general in period and locality. As a matter
of fact, it does appear with considerable regularity among the earliest
authentic historical names at Rome, as in Horatius Codes, Coriola-
nus, etc., particularly in the case of patrician families ; and it is not
too bold to suppose that the usage was even more general than the
records show. When we go outside of Latin, however, we find in
the Oscan and Umbrian dialects that the full name consisted oftenest
of praenomen and gentile. Only in a limited number of cases, where
Latin influence seems to have been at work, do we find instances of

1 82 George Davis Chase.

cognomina in these dialects. An examination of the proper names
in Zvetaieff, Inscrip. Ital. Infer., will show this to be the case.

In Latin we may look upon the three processes of abbreviation, of
restriction in number, and of addition of a second name as taking
place gradually and simultaneously, and each mutually tending to
produce the other. Many of the newly acquired cognomina did not
appear to the users radically different from the shortened praenomina;
and as many men came to be known and regularly called by their
cognomina, a number of the most frequent of these became confused
with the praenomina proper, and in time used in their stead. This
would explain such Greek names as IIAaT<ov and AmtxvAos, and in
Latin, praenomina like Gnaevos, Aulus, Proculus, etc., which must
have arisen as cognomina. In the case of most of these, their posi-
tion was precarious, and they never obtained a really permanent
foothold. We cannot suppose Agrippa, Proculus, Opiter, etc., ever
to have been in general and unrestricted use as praenomina, although
at times and in certain families they were certainly so employed.
At the same time their use as cognomina continued, and as such
they survived to classical times. In the other Italic dialects the
cognomina which arose at the earliest period seem in time to have
fully qualified as praenomina. The habit of forming new cognomina
then ceasing, all names were classed alike as praenomina. The
praenomina in Oscan appear to be partly like Gaius, Lucius, etc.,
partly like Gnaeus, Aulus, Proculus, etc. The three-name system in
Latin was due to the fact that cognomina never ceased to be formed,
even down to a late period.

The cognomina which finally took permanent station as praenomina
were Gnaeus, Aulus, Tiberius, and Spurius; for with regard to Titus,
Numerius, and Appius, which had their origin outside of Latium, we
are not able to decide to which class of names they belong.

Numerals could have arisen only as names of personal description.
Their existence as praenomina is very strong evidence that at an early
date the number of accepted praenomina was exceedingly limited.
The absence of the first four numerals and the frequent use of
Quintus and Sextus and occasional use of the higher ordinals show
that up to five a Roman was never forced to resort to numbering to
distinguish his sons, — that every pater familias had the free choice

The Origin of Roman Praenomina. 183

of at least four praenomina ; but that in the case of more than four
sons many were obliged to employ numerals. It does not follow
that every Roman had the same four praenomina to select from;
indeed we know that in many families certain common ones were
excluded (as in the case of Lucius among the Claudii, Suet. Tib. 1) ;
so that we may fairly consider the number of commonly accepted
praenomina in use at the early period at which the numerals were
added to have been six or seven. Those most universally in use
were Lucius, Gaius, Marcus, Publius ; to these were added often the
spurious praenomina, Gnaeus and Aulus; and Manius and Servius
were employed just enough to preserve their undisputed existence as
praenomina down to the latest times. All other names which we
have found occurring at one time or another as praenomina were
used among immigrants from other parts of Italy, or confined to
special families in such a way as not, to affect materially this ex-
tremely simple and impoverished system.

So far our explanation has taken no account of the gentilicium.
This did not arise in the same natural way as the other names. It
came about in a deliberate manner and in answer to the needs of a
highly developed political order. It arose with the formation of the
state, and was part of the official designation of the person. It must
have required a long time after the gentile came into official use
before it was regularly employed in familiar practice. It was only
at a late stage, though before the historical period, that a man was
regularly known by both praenomen and gentile.

At the time the gentile was introduced, the names in use were
single for each individual, and already of the two classes represented
by Lucius and Gnaeus. The gentile was placed always after these,
and thus its position as the second name was fixed. Here the devel-
opment stopped in Oscan and in Plebeian Latin. In Umbrian it was
the custom to express the father's name, in the genitive case, directly
after the praenomen, as C. V. Vistinie, Ner. T. Babr. (= C. Vistinii
V. f., Ner. Babrii T. f.) ; cf. Buecheler, Umbrica, p. 172. Here we
see most clearly the descriptive nature of the gentile ; a man was
known by his own name and his father's, but in addition- had a
political designation which was added as a matter of form. But
among the higher classes at Rome new cognomina continued to be

184 George Davis Chase.

added. These coming after the gentile made it more intimately a
part of the name, until in time it came to be the most distinctive
part in formal use. The old cognomina wavered and were divided
in use ; some few became permanently praenomina, but the majority,
after an interval of uncertainty, being recognized as belonging in
character to the new class, joined the cognomina, and are there
usually found.

The gentile was in all cases derived from already existing names.
The earliest ones were derived from names used as praenomina, as
Lucilius, Marcius, Procilius, Avilius, etc.; but as the number of
gentes continued to increase and the number of praenomina was
limited, they came to be formed from the later cognomina, as Longi-
lius, Luscius, Rubrius, Flavins, Calvius, and hosts of similar forms.

Thus roughly we have outlined the development of Roman names
down to the empire. At that time there began to creep in the great-
est confusion in names, owing to the common practice of multiplying
cognomina and the habit of leaving off the first names. But about
these later changes there is no mystery. The reason for them and
the course they took are both well established, and form no part of
the design of the present work.